High-resolution Remote Sensing Image Segmentation Using Minimum Spanning Tree Tessellation and RHMRF-FCM Algorithm

It is proposed a high resolution remote sensing image segmentation method which combines static minimum spanning tree(MST)tessellation considering shape information and the RHMRF-FCM algorithm.It solves the problems in the traditional pixel-based HMRF-FCM algorithm in which poor noise resistance and low precision segmentation in a complex boundary exist.By using the MST model and shape information,the object boundary and geometrical noise can be expressed and reduced respectively.Firstly,the static MST tessellation is employed for dividing the image domain into some sub-regions corresponding to the components of homogeneous regions needed to be segmented.Secondly,based on the tessellation results,the RHMRF model is built,and regulation terms considering the KL information and the information entropy are introduced into the FCM objective function.Finally,the partial differential method and Lagrange function are employed to calculate the parameters of the fuzzy objective function for obtaining the global optimal segmentation results.To verify the robustness and effectiveness of the proposed algorithm,the experiments are carried out with WorldView-3(WV-3)high resolution image.The results from proposed method with different parameters and comparing methods(multi-resolution method and watershed segmentation method in eCognition software)are analyzed qualitatively and quantitatively.

Application of High-Resolution Remote Sensing Technology in Quantitative Study on Coseismic Surface Rupture Zones: An Example of the 2008 M_w7.2 Yutian Earthquake

Objective Nowadays, high-resolution remote sensing technology has brought new changes to surveys of earthquakes, and the quantitative study of seismic faults based on this technology has become a trend in the world(Barzegari et al., 2017). An Mw 7.2 earthquake occurred in Yutian of Xinjiang on the western end of the Altyn Tagh fault on March 21 st, 2008. It is difficult to access this depopulated zone because of the high altitude and only 1–2 months of snowmelt. This study utilized high-resolution

A Remote Sensing Image Semantic Segmentation Method by Combining Deformable Convolution with Conditional Random Fields

Currently,deep convolutional neural networks have made great progress in the field of semantic segmentation.Because of the fixed convolution kernel geometry,standard convolution neural networks have been limited the ability to simulate geometric transformations.Therefore,a deformable convolution is introduced to enhance the adaptability of convolutional networks to spatial transformation.Considering that the deep convolutional neural networks cannot adequately segment the local objects at the output layer due to using the pooling layers in neural network architecture.To overcome this shortcoming,the rough prediction segmentation results of the neural network output layer will be processed by fully connected conditional random fields to improve the ability of image segmentation.The proposed method can easily be trained by end-to-end using standard backpropagation algorithms.Finally,the proposed method is tested on the ISPRS dataset.The results show that the proposed method can effectively overcome the influence of the complex structure of the segmentation object and obtain state-of-the-art accuracy on the ISPRS Vaihingen 2D semantic labeling dataset.

Recognizing and monitoring infectious sources of schistosomiasis by developing deep learning models with high-resolution remote sensing images

Background China is progressing towards the goal of schistosomiasis elimination,but there are still some problems,such as difficult management of infection source and snail control.This study aimed to develop deep learning models with high-resolution remote sensing images for recognizing and monitoring livestock bovine,which is an intermediate source of Schistosoma japonicum infection,and to evaluate the effectiveness of the models for real-world application.Methods The dataset of livestock bovine’s spatial distribution was collected from the Chinese National Platform for Common Geospatial Information Services.The high-resolution remote sensing images were further divided into training data,test data,and validation data for model development.Two recognition models based on deep learning methods(ENVINet5 and Mask R-CNN)were developed with reference to the training datasets.The performance of the developed models was evaluated by the performance metrics of precision,recall,and F1-score.Results A total of 50 typical image areas were selected,1125 bovine objectives were labeled by the ENVINet5 model and 1277 bovine objectives were labeled by the Mask R-CNN model.For the ENVINet5 model,a total of 1598 records of bovine distribution were recognized.The model precision and recall were 81.9%and 80.2%,respectively.The F1 score was 0.81.For the Mask R-CNN mode,1679 records of bovine objectives were identified.The model precision and recall were 87.3%and 85.2%,respectively.The F1 score was 0.87.When applying the developed models to real-world schistosomiasis-endemic regions,there were 63 bovine objectives in the original image,53 records were extracted using the ENVINet5 model,and 57 records were extracted using the Mask R-CNN model.The successful recognition ratios were 84.1%and 90.5%for the respectively developed models.Conclusion The ENVINet5 model is very feasible when the bovine distribution is low in structure with few samples.The Mask R-CNN model has a good framework design and runs highly efficiently.The livestock recognition models developed using deep learning methods with high-resolution remote sensing images accurately recognize the spatial distribution of livestock,which could enable precise control of schistosomiasis.

RepDDNet:a fast and accurate deforestation detection model with high-resolution remote sensing image

Forest is the largest carbon reservoir and carbon absorber on earth.Thus,mapping forest cover change accurately is of great significance to achieving the global carbon neutrality goal.Accurate forest change information could be acquired by deep learning methods using high-resolution remote sensing images.However,deforestation detection based on deep learning on a large-scale region with high-resolution images required huge computational resources.Therefore,there was an urgent need for a fast and accurate deforestation detection model.In this study,we proposed an interesting but effective re-parameterization deforestation detection model,named RepDDNet.Unlike other existing models designed for deforestation detection,the main feature of RepDDNet was its decoupling feature,which means that it allowed the multi-branch structure in the training stages to be converted into a plain structure in the inference stage,thus the computation efficiency can be significantly improved in the inference stage while maintaining the accuracy unchanged.A large-scale experiment was carried out in Ankang city with 2-meter high-resolution remote sensing images(the total area of it was over 20,000 square kilometers),and the result indicated that the model computation efficiency could be improved by nearly 30%compared with the model without re-parameterization.Additionally,compared with other lightweight models,RepDDNet also displayed a trade-off between accuracy and computation efficiency.

China’s high-resolution optical remote sensing satellites and their mapping applications

Since the beginning of the twenty-first century,several countries have made great efforts to develop space remote sensing for building a high-resolution earth observation system.Under the great attention of the government and the guidance of the major scientific and technological project of the high-resolution earth observation system,China has made continuous breakthroughs and progress in high-resolution remote sensing imaging technology.The development of domestic high-resolution remote sensing satellites shows a vigorous trend,and consequently,a relatively stable and perfect high-resolution earth observation system has been formed.The development of high-resolution remote sensing satellites has greatly promoted and enriched modern mapping technologies and methods.In this paper,the development status,along with mapping modes and applications of China’s high-resolution remote sensing satellites are reviewed,and the development trend in high-resolution earth observation system for global and ground control-free mapping is discussed,providing a reference for the subsequent development of high-resolution remote sensing satellites in China.

Advances in urban information extraction from high-resolution remote sensing imagery

The study of urban area is one of the hottest research topics in the field of remote sensing. With the accumulation of high-resolution(HR) remote sensing data and emerging of new satellite sensors, HR observation of urban areas has become increasingly possible, which provides us with more elaborate urban information. However, the strong heterogeneity in the spectral and spatial domain of HR imagery brings great challenges to urban remote sensing. In recent years, numerous approaches were proposed to deal with HR image interpretation over complex urban scenes, including a series of features from low level to high level, as well as state-of-the-art methods depicting not only the urban extent, but also the intra-urban variations. In this paper, we aim to summarize the major advances in HR urban remote sensing from the aspects of feature representation and information extraction. Moreover, the future trends are discussed from the perspectives of methodology, urban structure and pattern characterization, big data challenge, and global mapping.

Development and evaluation of low-altitude remote sensing systems for crop production management

Precision agriculture accounts for within-field variability for targeted treatment rather than uniform treatment of an entire field.It is built on agricultural mechanization and state-of-the-art technologies of geographical information systems(GIS),global positioning systems(GPS)and remote sensing,and is used to monitor soil,crop growth,weed infestation,insects,diseases,and water status in farm fields to provide data and information to guide agricultural management practices.Precision agriculture began with mapping of crop fields at different scales to support agricultural planning and decision making.With the development of variable-rate technology,precision agriculture focuses more on tactical actions in controlling variable-rate seeding,fertilizer and pesticide application,and irrigation in real-time or within the crop season instead of mapping a field in one crop season to make decisions for the next crop season.With the development of aerial variable-rate systems,low-altitude airborne systems can provide high-resolution data for prescription variable-rate operations.Airborne systems for multispectral imaging using a number of imaging sensors(cameras)were developed.Unmanned aerial vehicles(UAVs)provide a unique platform for remote sensing of crop fields at slow speeds and low-altitudes,and they are efficient and more flexible than manned agricultural airplanes,which often cannot provide images at both low altitude and low speed for capture of high-quality images.UAVs are also more universal in their applicability than agricultural aircraft since the latter are used only in specific regions.This study presents the low-altitude remote sensing systems developed for detection of crop stress caused by multiple factors.UAVs,as a special platform,were discussed for crop sensing based on the researchers'studies.

Monitoring of vegetation coverage based on high-resolution images

Measurement of vegetation coverage on a small scale is the foundation for the monitoring of changes in vegetation coverage and of the inversion model of monitoring vegetation coverage on a large scale by remote sensing. Using the object-oriented analytical software, Definiens Professional 5, a new method for calculating vegetation coverage based on high-resolution images （aerial photographs or near-surface photography） is proposed. Our research supplies references to remote sensing measurements of vegetation coverage on a small scale and accurate fundamental data for the inversion model of vegetation coverage on a large and intermediate scale to improve the accuracy of remote sensing monitoring of changes in vegetation coverage.

Estimation of carbon and nitrogen contents in citrus canopy by low-altitude remote sensing

The study aimed to investigate the fast and nondestructive method for detecting carbon and nitrogen content in citrus canopy.The multispectral imagery of Tarocco blood orange(Citrus sinensis L.Osbeck)plant canopy was obtained by a multispectral camera array mounted at an eight-rotor unmanned aerial vehicle(UAV)flying at an altitude of 100 m above the canopy in Wanzhou District of Chongqing Municipality,China.Average spectral reflectance data of the whole canopy,mature leaf areas and young leaves areas were extracted from the imagery.Two spectral pre-processing methods,multiplicative scatter correction(MSC)and standard normal variable(SNV),and two modeling methods,the partial least squares(PLS)and the least squares support vector machine(LS-SVM),were adopted and compared for their prediction accuracy of total content of nitrogen,soluble sugar and starch in the leaves.The results showed that,based on the spectral data extracted from the mature leaves in the multispectral imagery,the PLS model based on the original spectrum obtained a Rp(correlation coefficient)of 0.6469 and RMSEP(root mean squares error of prediction)of 0.1296,suggested that it was the best for the prediction of total nitrogen content;the PLS model based on MSC(multiplicative scatter correction)spectrum pre-processing was the best for predicting total soluble sugar content(Rp=0.6398 and RMSEP=8.8891);and the LS-SVM model based on MSC was the best for the starch content prediction(Rp=0.6822 and RMSEP=14.9303).The prediction accuracy for carbon and nitrogen contents based on the spectral data extracted from the whole canopy and the young leaves were lower than that from the mature leaves.The results indicate that it is feasible to estimate the carbon and nitrogen contents by low-altitude airborne multispectral images.

Monitoring the green evolution of vernacular buildings based on deep learning and multi-temporal remote sensing images

The increasingly mature computer vision(CV)technology represented by convolutional neural networks(CNN)and available high-resolution remote sensing images(HR-RSIs)provide opportunities to accurately measure the evolution of natural and artificial environments on Earth at a large scale.Based on the advanced CNN method high-resolution net(HRNet)and multi-temporal HR-RSIs,a framework is proposed for monitoring a green evolution of courtyard buildings characterized by their courtyards being roofed(CBR).The proposed framework consists of an expert module focusing on scenes analysis,a CV module for automatic detection,an evaluation module containing thresholds,and an output module for data analysis.Based on this,the changes in the adoption of different CBR technologies(CBRTs),including light-translucent CBRTs(LT-CBRTs)and non-lighttranslucent CBRTs(NLT-CBRTs),in 24 villages in southern Hebei were identified from 2007 to 2021.The evolution of CBRTs was featured as an inverse S-curve,and differences were found in their evolution stage,adoption ratio,and development speed for different villages.LT-CBRTs are the dominant type but are being replaced and surpassed by NLT-CBRTs in some villages,characterizing different preferences for the technology type of villages.The proposed research framework provides a reference for the evolution monitoring of vernacular buildings,and the identified evolution laws enable to trace and predict the adoption of different CBRTs in a particular village.This work lays a foundation for future exploration of the occurrence and development mechanism of the CBR phenomenon and provides an important reference for the optimization and promotion of CBRTs.

A NEW METHOD FOR IMPROVING THE ABILITY OF ATMOSPHERIC INFRARED REMOTE SENSING FROM THE GROUND

Research of the inversion method of atmospheric parameter is a very important aspect for extracting more information from measured data. Some authors have done a lot of re- searches in this field. They mainly searched for the way to obtain atmospheric parameter characteristics effectively from the measured information. Although Rodgerst pointed out that the statistical method using some a priori information could greatly improve the

Integration of optical and SAR remote sensing images for crop-type mapping based on a novel object-oriented feature selection method

Remote sensing is an important technical means to investigate land resources.Optical imagery has been widely used in crop classification and can show changes in moisture and chlorophyll content in crop leaves,whereas synthetic aperture radar(SAR)imagery is sensitive to changes in growth states and morphological structures.Crop-type mapping with a single type of imagery sometimes has unsatisfactory precision,so providing precise spatiotemporal information on crop type at a local scale for agricultural applications is difficult.To explore the abilities of combining optical and SAR images and to solve the problem of inaccurate spatial information for land parcels,a new method is proposed in this paper to improve crop-type identification accuracy.Multifeatures were derived from the full polarimetric SAR data(GaoFen-3)and a high-resolution optical image(GaoFen-2),and the farmland parcels used as the basic for object-oriented classification were obtained from the GaoFen-2 image using optimal scale segmentation.A novel feature subset selection method based on within-class aggregation and between-class scatter(WA-BS)is proposed to extract the optimal feature subset.Finally,crop-type mapping was produced by a support vector machine(SVM)classifier.The results showed that the proposed method achieved good classification results with an overall accuracy of 89.50%,which is better than the crop classification results derived from SAR-based segmentation.Compared with the ReliefF,mRMR and LeastC feature selection algorithms,the WA-BS algorithm can effectively remove redundant features that are strongly correlated and obtain a high classification accuracy via the obtained optimal feature subset.This study shows that the accuracy of crop-type mapping in an area with multiple cropping patterns can be improved by the combination of optical and SAR remote sensing images.

Aquaculture area extraction and vulnerability assessment in Sanduao based on richer convolutional features network model

Sanduao is an important sea-breeding bay in Fujian,South China and holds a high economic status in aquaculture.Quickly and accurately obtaining information including the distribution area,quantity,and aquaculture area is important for breeding area planning,production value estimation,ecological survey,and storm surge prevention.However,as the aquaculture area expands,the seawater background becomes increasingly complex and spectral characteristics differ dramatically,making it difficult to determine the aquaculture area.In this study,we used a high-resolution remote-sensing satellite GF-2 image to introduce a deep-learning Richer Convolutional Features(RCF)network model to extract the aquaculture area.Then we used the density of aquaculture as an assessment index to assess the vulnerability of aquaculture areas in Sanduao.The results demonstrate that this method does not require land and water separation of the area in advance,and good extraction can be achieved in the areas with more sediment and waves,with an extraction accuracy>93%,which is suitable for large-scale aquaculture area extraction.Vulnerability assessment results indicate that the density of aquaculture in the eastern part of Sanduao is considerably high,reaching a higher vulnerability level than other parts.

Overview of the key technologies for high-resolution satellite mapping

As the important infrastructures for land mapping and resource monitoring,highresolution remote sensing satellites(HRSS)are urgently demanded for the development of China.In this article,the key technologies of the main HRSS are summarized,and these technologies include sensor design,attitude and orbit determination,geometric calibration,imaging model construction,and block adjustment,etc.,which involve the mapping accuracy of HRSS.Finally,the system design of the ZY-3 Satellite(China’s first civil stereoscopic surveying and mapping satellite,to be launched in 2012)is introduced,which mainly include satellite technical specifications and strategies design based on these key technologies research.

Assessment and Modeling of Geo-Spatial Technology and Geo-Spatial Intelligence Support for Joint Military Operations

World military force structure is dramatically changing as collectively;our armed forces undergo a major transition from unprofessional to the Objective Force (designed to capitalize on information-age based technologies and Human Interaction to Non-Human Interaction). Traditional “stovepipes” among services are being eliminated and replaced with integrated systems that allow joint forces (combined Army, Air Force and navy) to seamlessly execute required tasks. This study was undertaken in conjunction with Geospatial Technology (Shows Space and Time) and Geospatial Intelligence Analysis (Use Algorithm, Use AI Concepts, IMINT and GEOINT). In order to successfully support current and future Ethiopian military operations in war zones, geospatial technologies and geospatial intelligence must be integrated to accommodate force structure evolution and mission requirement directives. The intent of joint intelligence operations is to integrate Ground, Air and Navy Forces at war zone and also give COP (“common operational picture”) for Operational and Tactical Commander Service and national intelligence capabilities into a unified effort that surpasses any single organizational effort and provides the most accurate and timely intelligence to commanders.

UAV-based spatial pattern of three-dimensional green volume and its influencing factors in Lingang New City in Shanghai,China

Three-dimensional green volume(TDGV)reflects the quality and quantity of urban green space and its provision of ecosystem services;therefore,its spatial pattern and the underlying influential factors play important roles in urban planning and management.However,little is known about the factors contributing to the spatial pattern of TDGV.In this paper,TDGV and land use intensity(LUI)extracted from high spatial resolution(0.05 m)remotely sensed data acquired by an unmanned aerial vehicle(UAV),anthropogenic factors^(1))and natural factors^(2))were utilized to identify the spatial pattern of TDGV and the potential influencing factors in Lingang New City,a rapidly developed coastal town in Shanghai.The results showed that most of the TDGV was distributed in the western part of this new city and that its spatial variations were significantly axial.TDGV corresponded well with the chronologies of land formation,urban planning,and construction in the city.Generalized least squares(GLS)analysis of TDGV(grid cell size:100×100 m)and its influencing factors showed that the TDGV in this new city was significantly negatively correlated with both LUI and distance from roads and significantly positively correlated with land formation time and distance from water.Distance from buildings did not affect TDGV.Additionally,the degree of influence decreased in the following order:distance from water>land formation time>distance from roads>LUI.These results indicate that the spatial pattern of TDGV in this new town was mainly affected by natural factors(i.e.,the distance from water and land formation time)and that the artificial disturbances caused by rapid urbanization did not decrease the regional TDGV.The main factors shaping the spatial distribution of TDGV in this city were local natural factors.Our findings suggest that the improvement in local soil and water conditions should be emphasized in the construction of new cities in coastal areas to ensure the efficient provision of ecological services by urban green spaces.

Mapping urban building stocks for vulnerability assessment
preliminary results

This paper discusses a methodology to collect building inventory data by combining image processing techniques,field work or tools such as Google Street View and applying statistical inferences.Following the methodology outlined in Marinescu(2002),a family of Gabor filters are first constructed,which are then applied to an optical high-resolution image.The output from the processed image is segmented using Self-Organising Maps.This paper examines the relationship between the segmented areas in the image and the building type distribution within each segmented area,by deriving the distribution from field data.The relationship between the average number of buildings in these cells against the number of grid cells allocated to each segmentation cluster is also investigated.Finally,using these results,the overall building inventory distribution for the whole of the case study site of Pylos is presented.