Automatic road extraction has gained significant attention in urban navigation, sustainable transport, and disaster response. Conventional convolutional neural networks (CNNs) operate within the local receptive field, limiting their capacity to represent potential global relations between roads and surroundings. In addition, the edge is important topological information for road targets. Several works focus on predicting precise boundaries to enhance road extraction. However, over fit edges and the course integration between features of different network layers may lead to loss of local details and incorrect road segmentation results. Therefore, the Road-detail Preserving Mapper (RoadDP-Mapper) framework is proposed. First, RoadDP-Mapper employs a hierarchical transformer as the encoder to enable local-to-global reasoning. The asymmetric upsampling layers (APLs) are introduced to enhance the model’s capability to perceive and reconstruct critical road detail information. Second, a road edge-constrained branch with a detail preservation module (DPM) is devised to amplify the distinction between roads and backgrounds by extracting and preserving explicit class boundary details. The proposed joint loss inspires the transformer to capture the contextual spatial relationships while preserving the fine-grained features of the road. We evaluated our framework on the DeepGlobe dataset and self-annotated images from ten representative cities in China. The proposed framework has demonstrated its effectiveness by significantly reducing both missed detections and false alarms in road extraction. Furthermore, spatial transfer experiments have confirmed the generalizability of RoadDP-Mapper for large-scale road mapping.

Accurately determining the spatial position and distribution structure of agricultural cultivation parcels (ACPs) is essential for regional agricultural planning and food security. Currently, utilizing deep learning technology based on very high resolution remote sensing imagery has proven effective for intelligent parcel extraction. However, relying solely on the model output, especially from single-task models in mountainous regions with complex, heterogeneous, and fragmented smallholder agriculture, remains questionable. To address this challenge, leveraging geographical prior knowledge is critical. This article proposes using the deep semantic segmentation algorithm in conjunction with comprehensive prior strategies. An improved densely connected link network (D-LinkNet) is employed to delineate the parcels, while geographical zoning, coarse spatial scope, stratification strategy, and homogeneity checking are exerted to understand regions, facilitate samples, reduce interferences, decompose objects, and identify undersegmentation. The proposed framework was validated in Jiangjin district, Chongqing of China, using Gaofen-2 images as the vital data. Compared to the method relying solely on deep learning, our method achieved superior performance with an overall accuracy of 0.924, Kappa coefficient of 0.847,$F1$score of 0.921, and IoU exceeding 0.8. Moreover, the results demonstrated high accuracy in the individual geometric precision of parcel. Over 1.23 million parcels were identified, comprising 77% cultivated land and 23% garden land. The areal proportion of paddy fields, drylands, and pepper gardens approximated 1:1:1, consistent with statistical data. This method offers a feasible approach for finely extracting agricultural parcels.

For large-scale mapping applications, cross-domain hyperspectral image classification (HSIC) has emerged as a highly promising research area. However, the classification accuracy decreased significantly when unseen classes emerged. Few shot learning (FSL) methods are adopted in cross-domain HSIC methods to address this problem. Despite this, existing cross-domain HSIC methods still have three key issues that hamper their classification capabilities: 1) previous works struggle to balance incorporating distinctive intradomain knowledge and managing model complexity in the face of significant domain representation differences; 2) previous works inadequately consider the limited capture capacity of interdomain intrinsic mutually invariant structures; and 3) previous works fail to capture the distinct characteristics of both head categories (e.g., urban buildings) and tail categories (e.g., urban corn) simultaneously when applying FSL to deal with unseen classes problem. In this article, we propose a cycle-resemblance few-shot transformation (CF-Trans) network to effectively handle the aforementioned challenges by integrating intradomain distinctiveness with interdomain invariance. To facilitate efficient intradomain feature aggregation for HSI, a novel lightweight intradomain attentive network is introduced. Different from previous works, to reduce the negative impact caused by inaccurate classifier predictions, from the perspective of interdomain knowledge transformation, a cycle-resemblance adversarial network is designed to capture the intrinsic mutually invariant structures. A dynamic label expansion mechanism is designed to capture the distinctive intradomain features of the head and tail classes. Experimental results on six HSI datasets including agricultural, rural-urban and urban datasets show the remarkably performance of our network.

To address the limitations of hyperspectral imaging systems, super-resolution (SR) techniques that fuse low-resolution hyperspectral image (HSI) with high-resolution multispectral image (MSI) are applied. Due to the significant modal difference between HSI and MSI, and the insufficient consideration of HSI band correlation in previous works, issues arise such as spectral distortions and loss of fine texture and boundaries. In this article, an unsupervised spectral correlation-based fusion network (SCFN) is proposed to address the above challenges. A new dense spectral convolution module (DSCM) is proposed to capture the intrinsic similarity dependence between spectral bands in HSI to effectively extract spectral domain features and mitigate spectral aberrations. To preserve the rich texture details in MSI, a global-local aware block (GAB) is designed for joint global contextual information and emphasize critical regions. To address the cross-modal disparity problem, new joint losses are constructed to improve the preservation of high-frequency information during image reconstruction and effectively minimize spectral disparity for more precise and accurate image reconstruction. The experimental results on three hyperspectral remote sensing datasets demonstrate that SCFN outperforms other methods in both qualitative and quantitative comparisons. Results from the fusion of real hyperspectral and multispectral remote sensing data further confirm the applicability and effectiveness of the proposed network.

Recently, high-resolution (HR) remote sensing images have attracted increasing attention in a number of tasks. Super-resolution (SR) is an efficient method to obtain high-resolution remote sensing images. Due to the influence of imaging distances and angles, remote sensing images significantly differ from natural images in terms of land cover element distribution, ground object scale and scene complexity. This poses a challenge for capturing global and local low- and high-frequency and restoring fine image details for remote sensing image SR. In this article, a local-global context-aware generative dual-region adversarial network (LGC-GDAN) is designed for remote sensing image SR. It is composed of dual region-level discriminators and a dual-path generator with a context-aware network and an edge-assisted network. To capture global and local low- and high-frequency information, the global-aware self-attention (GAS) mechanism and local-aware self-attention (LAS) mechanism are introduced into the context-aware network. The GAS mechanism combines high-pass and low-pass filtering for long-range similarity feature, while LAS uses local aggregation for fine-level feature. The LR images and the corresponding edge maps are input to the edge-assisted network to extract the detailed geometric structure. To address small ground object and complex ground scenes, conventional image-level discriminators exhibit limited performance in capturing detailed information. Unlike previous discriminator, a region-level discriminator is designed to obtain the real/fake label of each local region. Moreover, two task-driven loss functions are designed to produce diverse images for further scene classification. The experiments undertaken on several remote sensing datasets demonstrate that LGC-GDAN outperforms the other state-of-the-art methods.

Because of the primary role of land surface temperature (LST) in the physical processes of surface energy balance at local through global scales, dynamic, continuous, and seamless LST monitoring is constantly in urgent need. Thermal infrared (TIR) remote sensing serves as the most commonly used sources for LST retrieval owing to its relatively fine spatial-temporal resolution and presentable accuracy. However, limited by the inability to penetrate clouds, original TIR LST data suffers significantly from data missing problems. Furthermore, the view time of pixels along the scan line differs significantly for polar-orbiting satellites, exerting appreciable influence on the subsequent data applications. To cope with the above setbacks simultaneously, we proposed a practical reconstruction framework based on the inner physical connection between LST and solar radiation, which was accurately expressed by random forest regression model, with the consideration of various auxiliary environmental factors (i.e., elevation, slope, longitude, latitude, and surface reflectance). Taking the Tibetan Plateau (TP) as the study area, the proposed method was applied to generate spatially seamless and time-consistent LST products with the use of the Moderate Resolution Imaging Spectroradiometer (MODIS) Terra daytime LST product. From visual assessment, the reconstructed product exhibits ideal spatial-temporal continuity within the TP. Through the validation with in-situ observations from five different stations, the results show a higher consistency with ground measurements than the LST product from the Global Land Data Assimilation System (GLDAS) and other all-weather LST product, with an average improvement on RMSE of 1.06 K and 1.59 K under clear conditions, and 1.86 K and 2.72 K under cloudy conditions. The validation demonstrates that the proposed method is well applicable for all-weather LST reconstruction over a large-scale area with significant surface heterogeneity, which also shows good ability to remove the temporal inconsistency induced by satellite observations. Additionally, it can be reliably generalized to different areas with similar data requirements for its sufficient effectiveness and flexibility.

The formation of shadows in very high spatial resolution (VHR) remote sensing imagery is attributed to light being blocked by objects, reducing spectral radiance in the shadow landscape. An accurate and robust shadow removal method can recover spectral and textural information and, hence, is a crucial preprocessing step for urban image analyses. In this study, we develop a KnOwledge-driven shadow progressive removal (KO-Shadow) framework with three subnets for VHR imagery using a weakly supervised manner. Specifically, the shadow preelimination subnet is proposed to initially address the large chromatic aberration between the real and shadow situations. Then, the prior knowledge-guided refinement subnet is proposed to refine the preelimination results by mining tone and texture information. Moreover, the locality feature discriminator is designed for region-specific evaluation of the generated shadow-free samples to improve the capacity of subnets. Experimental results of six typical cities in the world show that KO-Shadow is superior to the existing methods. Moreover, the generalizability analysis in complex urban scenarios validates the robustness of our method. The shadow recovery score (SRI) is proposed to evaluate the spectral similarities between the recovered area and shadow-related land-cover types (e.g., road, building, and lawn). The results show that KO-Shadow can yield more visually realistic shadow-free images and better quantitative performance. Overall, KO-Shadow provides a new perspective for VHR image shadow removal by mining the prior knowledge of the complex shadows in urban areas.

BACKGROUND: Clustering analysis is widely used to interpret biomedical data and uncover new knowledge and patterns. However, conventional clustering methods are not effective when dealing with sparse biomedical data. To overcome this limitation, we propose a hierarchical clustering method called polynomial weight-adjusted sparse clustering (PWSC). RESULTS: The PWSC algorithm adjusts feature weights using a polynomial function, redefines the distances between samples, and performs hierarchical clustering analysis based on these adjusted distances. Additionally, we incorporate a consensus clustering approach to determine the optimal number of classifications. This consensus approach utilizes relative change in the cumulative distribution function to identify the best number of clusters, resulting in more stable clustering results. Leveraging the PWSC algorithm, we successfully classified a cohort of gastric cancer patients, enabling categorization of patients carrying different types of altered genes. Further evaluation using Entropy showed a significant improvement (p = 2.905e-05), while using the Calinski-Harabasz index demonstrates a remarkable 100% improvement in the quality of the best classification compared to conventional algorithms. Similarly, significantly increased entropy (p = 0.0336) and comparable CHI, were observed when classifying another colorectal cancer cohort with microbial abundance. The above attempts in cancer subtyping demonstrate that PWSC is highly applicable to different types of biomedical data. To facilitate its application, we have developed a user-friendly tool that implements the PWSC algorithm, which canbe accessed at http://pwsc.aiyimed.com/ . CONCLUSIONS: PWSC addresses the limitations of conventional approaches when clustering sparse biomedical data. By adjusting feature weights and employing consensus clustering, we achieve improved clustering results compared to conventional methods. The PWSC algorithm provides a valuable tool for researchers in the field, enabling more accurate and stable clustering analysis. Its application can enhance our understanding of complex biological systems and contribute to advancements in various biomedical disciplines.

Missing data reconstruction is a classical yet challenging problem in remote sensing images. Most current methods based on traditional convolutional neural network require supplementary data and can only handle one specific task. To address these limitations, we propose a novel generative adversarial network-based missing data reconstruction method in this letter, which is capable of various reconstruction tasks given only single source data as input. Two auxiliary patch-based discriminators are deployed to impose additional constraints on the local and global regions, respectively. In order to better fit the nature of remote sensing images, we introduce special convolutions and attention mechanism in a two-stage generator, thereby benefiting the tradeoff between accuracy and efficiency. Combining with perceptual and multiscale adversarial losses, the proposed model can produce coherent structure with better details. Qualitative and quantitative experiments demonstrate the uncompromising performance of the proposed model against multisource methods in generating visually plausible reconstruction results. Moreover, further exploration shows a promising way for the proposed model to utilize spatio-spectral-temporal information. The codes and models are available athttps://github.com/Oliiveralien/Inpainting-on-RSI.

The spatial downscaling of soil moisture (SM) provides a technical tool to solve the problem of coarse resolution of passive microwave products. However, conventional methods are developed based on the kilometer scale grid pixels of remote sensing images. The regular rough grids will lead to the mixing and uncertainty of SM information. In this paper, we formulate a novel land geoparcel-based spatial downscaling technique for the Soil Moisture Active Passive (SMAP) satellite products. It is developed by combining XGBoost (eXtreme Gradient Boosting) machine learning algorithm with the support of geoparcel vector data and a variety of auxiliary raster data. The downscaling effect is evaluated by using SMAP 9km products and site measured data in Tongnan District of Chongqing, China. The experiments show that the geoparcel-based downscaling method maintains the dynamic range of the original SM product, and conserves energy before and after downscaling. It is proved that our method effectively increases the spatial details of the original SM product with complete spatial coverage. The comparison and analysis with the ground verification data demonstrate that the formalized procedure with geoparcel-based spatial downscaling allows better results than those of using km-scale regular grids.

Tumor mutational burden (TMB) is the most important and most promising biomarker in the era of tumor immunotherapy, and it can predict the immunotherapy efficiency of patients in various cancers including liver cancer. TMB is mainly obtained by next generation sequencing technology such as whole exome sequencing (WES). However, conditions such as excessive testing costs, lengthy detection cycles, and tissue sample dependence severely limit the clinical application of TMB. Inspired by the inner link between the intrinsic characteristics of the tumor cell genome and the pathological features of tumor cells and their microenvironment-related cells, we propose a deep learning method for predicting the level of TMB (high or low) directly from pathological images. This study found that the feature scale (receptive field) is the biggest factor affecting the classification effect of TMB prediction, and further determined the best receptive field through a series of experiments. Experimental results show that our method is far more out performance of the commonly used panel sequencing (99.7% VS 79.2%). To the best of our knowledge, this is the first research to predict TMB and the highest level of accuracy of genomic characteristic predicted by pathological images. The proposed method has the potential to provide immunotherapy to a much broader subset of patients with liver cancer.

Object-based change detection (CD) is an effective method of identifying detailed changes in land features by contrastively observing the same areas of high-resolution remote sensing images at different times. Binarization is the important step in partitioning changed and unchanged classes in the unsupervised domain. We formulate a novel binarization technique based on the Weibull mixture model, where generated similarity measure images are modeled using a mixture of nonnormal Weibull distributions. The parameters in the model are further globally estimated by employing a genetic algorithm. Two data sets with high-resolution remote sensing images are used to evaluate the effectiveness of the proposed method. Experimental results demonstrate that the method allows better and more robust unsupervised object-based CD than do state-of-the-art threshold-based and clustering-based methods. Advantages of the proposed method are embodied in the modeling of relatively few data of the changed class with a skewed and long tail distribution.

Urban water extraction from high-resolution remote sensing image is one of important aspects for regional-urban environment research. However, the past researches mainly centered upon moderate and low resolution image, water body in open country and global scale for water extraction, which can't achieve better performance. To solve those problems, an adaptive method to extract urban water was proposed based on local iteration using high-resolution remote sensing image. In each iteration, segmented buffers were constructed with adaptive length and radius to exploit information in local scale, and also spatial consistency was token into account for an improvement of local classification. Experiments demonstrated that the proposed method was applicable in urban water extraction from high-resolution image.

Impervious surface percentage(ISP) is the key parameter for urban regional environment research. This paper proposes the method of ISP estimation by using support vector machine(SVM) on TM image: (1) extract the ISA pixels which occupies any portion of the constructed impervious class based on SVM classification for spatial inputs of ISP estimation (2) estimate ISP of ISA pixels by using SVM regression model, build sample-ISP regression model based on various spectral features inputs and apply ISP-model for regional imperviousness mapping. On the TM image of Tianjin urban area, select high resolution image(Quickbird) classification result of college, industrial and residential districts as training sample(7500 items) and testing sample(2000 items), the mean square error(RMSE) of SVM model is 15.4%; adding “greenness” of tasseled cap transform as SVM feature, the RMSE decrease to 12%. The results of the study indicate that SVM model is suitable for large area ISP mapping without insufficient sample because of the non-linear characteristic and good performance of small-sample generalization. Additionally, to build a typical sample library for large-area ISP mapping will be our future research directions.

Remote sensing information extraction is the key step of remote sensing application, and the automatic and high-precise extraction of water information from remotely sensed images is of great significance and urgently required in many research fields. This paper presents a step-by-step iterative transformation mechanism to extract water information, which uses spatial scale transformation mechanism of “whole-local” based on water index fitted from spectral library using spectral angle method first, and then fuses the hierarchical knowledge of water extraction and achieves the gradually approach of the water body's optimal margin iteratively by combining the segmentation and classification at whole and local scales respectively. Experiment of plateau lake information extraction demonstrates its better accuracy and efficiency.

Tibetan Plateau is a typical study area of global environmental change, and lake is an important ecological factor to reveal eco-environmental evolution. Using remote sensing technology to monitor the succession law of lakes on the plateau is of great significance to global environment change research. Based on water index computed by spectral feature fitting (SFF) method, this paper uses “whole-local ” spatial scale transformation mechanism, along with iterative algorithm, to obtain high-precise extraction of modern lakes on the Tibetan Plateau. Moreover, uses integrated data of LANDSAT ETM+ images and SRTM data to further detect and recover paleo shorelines. By comparing paleo and modern lakes, it shows that lakes on the Tibetan Plateau have shrunk significantly since the great lake period, which provides fundamental information support to researches on global paleo-climatology and paleo-hydrology change.

As a main factor of evaluating city development speed, road is one of the fast information updating element during city development. Road information extraction based on high-resolution remotely sensed images has very important significance because road affects city land use-cover change. Based on the synthetically analyzing all kinds of read extraction method from high-resolution remotely sensed images, this paper presents the road extraction method that firstly extracts parcel-units from the experimental image based on different scales, and then analyzes the parcel-units assisted by transcendental knowledge. This method can extract road information exactly by object-oriented method, and can acquire different-scale roads according to the user's requirement. At last this paper gives the road information extraction result analysis of the experiment sample.

Edge based image registration is efficient for multisensor image registration. However, original similarity measures are sensitive to noise or geometric distortion. In order to solve the problems, a new similarity measure is proposed combining chord distance and local geometric modal. Meanwhile consistency between global and local geometric modal is used to enhance the robustness of registration. Only matched edge segments consistent with both global and local geometric modal are reserved as final matched results. Experiments show that the proposed method is applicable in registering multisensor images with strong edges.

Synthetic variable ratio (SVR) method was first introduced to fuse panchromatic (PAN) image and multi-spectral (MS) image by Muechika et al. in 1993, and was improved by Zhang Y. in 1999 and 2001 respectively. As for Muechika SVR method, it isn't suitable for fusing PAN and MS which cover large area. And for Zhang Y. SVR(ZY-SVR) method, on the one hand it needs to select a great number of pixels of different land coverage classes to conduct multiple regression analysis and thus it seems greatly empirical; on the other hand the coefficients obtained through regression lack physical meanings. This paper puts forward a new method called LAB-SRV which introduces the sensor spectral response into SVR method using the CIELab color space. Several experiments done in this paper indicate that this method can sharpen multi-spectral MS images without changing their spectral characters very much and have an advantage over ZY-SVR method with PAN and MS of IKONOS and QuickBird.

According to the characteristics of urban spatial information, this paper presents a multi-scale urban spatial information extraction mode from high-resolution remote sensing images based on feature units. Scale-transforming technique is emphasized in this mode. By rough-classification using region partition method based on GMRF-SVM, we can get relatively great target areas and extract water body first. And then, using block parcel unit extraction method based on histogram threshold segmentation and linear parcel unit extraction method based on edge detection, we obtain urban construction area, road area and vegetation area respectively. Based on the extraction mode of feature units, a set of methods are built up to extract urban spatial information from high-resolution remote sensing images. At last, this paper uses QuickBird image, and chooses a typical urban area as the test data. The image is classified as water body, road, green space and construction area, and we use landscape models to further analyze spatial information of urban land use.

This paper analyses the advantages of remote sensing image processing per-parcel compared to per-pixel; Based on huaman's visual mechanism and theory of scale spatial, this paper designs the technique flow of multi-scale information extraction from high resolution remote sensing images based on features: rough classification - parcel unit extraction expression of feature - ntelligent illation - information extraction or target recognition. Especially, this paper expatiates the work flow of block and linear parcel unit extraction in detail and gives tests with IKONOS and QUICKBIRD images. The test shows that the methods of this paper is convenient to integrating visual and environmental knowledge and can improve the level of automatization and intelligentization of remote sensing data process and application.

Clustering is a widely used technique of finding interesting patterns residing in the dataset that were not obviously known. It is a division of data into groups of similar objects. The clustering of large data sets has received a lot of attention in recent years, however, clustering is still a challenging task since many cluster algorithms fail to do well in scaling with the size of the data set and the number of dimensions that describe the points, or in finding arbitrary shapes of clusters, or dealing effectively with the presence of noise. This paper describes a clustering method for unsupervised classification of objects in large data sets. The new methodology combines the simulating annealing algorithm with CLARANS (clustering large application based upon randomized search) in order to cluster large data sets efficiently. At last, the method is experimented on the generated data set. The result shows that the approach is quick than CLARANS and can produce a similar division of data as CLARANS.

The general image segmentation approach used in other domains may not be applicable to the remote sensing field, which is due to the following factors: remotely sensed data is multi-spectral, always very large in size, and in multi-scale as well. How to quickly and efficiently segment remotely sensed imagery is still a big issue to be solved. Based on human vision mechanism, a new hybrid multi-scale segmentation approach is presented, which is implemented at three coarse-to-fine scale levels. First, remotely sensed imagery is segmented at a coarse scale, and image regions (segments) are produced. Then, the corresponding regions in the original image are segmented by another segmentation approach one by one at the fine scale. From the experiment results, we found the approach is rather promising. However, there still exists some problems to be settled, and further researches should be conducted in the future.

An elliptical basis function (EBF) network is proposed in this study for the classification of remotely sensed images. Though similar in structure, the EBF network differs from the well-known radial basis function (RBF) network by incorporating full covariance matrices and uses the expectation-maximization (EM) algorithm to estimate the basis functions. Since remotely sensed data often take on mixture-density distributions in the feature space, the proposed network not only possesses the advantage of the RBF mechanism but also utilizes the EM algorithm to compute the maximum likelihood estimates of the mean vectors and covariance matrices of a Gaussian mixture distribution in the training phase. Experimental results show that the EM-based EBF network is faster in training, more accurate, and simpler in structure.