Shoulder line is one of the most distinguishing terrain structure lines in Loess Plateau. However, in the loess hilly area, the point cloud near the shoulder line is usually erased mistakenly while using the unified vegetation removal algorithm, which could result in the decrease of DEM accuracy in the terrain expression. Thus the shoulder line should be extracted firstly before removing the vegetation. This paper has proposed a novel shoulder line extraction method based on point cloud data using the multi-scale sampling and slope threshold segmentation. Firstly, a surface model was built based on the ground points selected by a proper filter window. Then, based on the constructed surface model, the slope was calculated and a significant difference near the shoulder line would emerge. Finally, the shoulder line was generated using the slope threshold segmentation. Through the multi-iterative experiments, the optimal filter sizes for each of the 8 sample areas were found, and a power function between the filter size and the point density was discovered. The correctness and reasonability of this correlation was verified in another 5 test areas. Then, the shoulder line of the whole area was generated by using this verified correlation. Using the overlay analysis over the manual identification result, the extracted shoulder line has an accuracy of 85% (within a 0.5 m buffer area). As a conclusion, this method could contribute to improve the accuracy of vegetation removal algorithm that uses the point cloud data for the loess hilly area.
Building facade is an important component of urban street features. Delineating and representing the building facade would benefit the urban building design and planning. As a new mobile mapping system, Mobile Laser Scanning (MLS) allows the quick and cost-effective acquisition of close-range three-dimensional (3D) measurements of urban street objects. This paper presents a semiautomated segmentation method for identifying the building facades from MLS point clouds data. The method consists of three major steps: (1) a horizontal grid system is built for the study area, and the multidimensional geometric features of 3D point clouds data, including the normal vector feature, omni-variance feature, geometric dimensionality of α1, α2 and α3, and eigen-entropy feature, are defined and calculated. Then, a feature image is created after projecting these features to the horizontal grid. (2) Building facades are roughly extracted using Support Vector Machine (SVM). (3) The rough extraction result is filtered according to the characteristics of grid including the shape coefficient, grid′s area, and the largest elevation. Two MLS point cloud datasets of Carnegie Mellon University (CMU) database were used in this study to estimate the feasibility and effectiveness of the method. It was found that this method performs well in extracting the building facades. The precision of the results is 0.88, and its recall rate is 0.90, which is better than some existing methods. Our method provides an effective tool for extracting building facades of streets from MLS point cloud data.
Micro-blogs usually contain abundant types of geographical event information, which could compensate for the shortcomings of traditional fixed point monitoring technologies and improve the quality of emergency response. Identify the micro-blog messages that containing the geographical event information is the prerequisite for fully utilizing this data source. The trigger-based and the supervised machine learning methods are commonly adopted to identify the event related texts. Comparatively, the supervised machine learning methods have better performance than the trigger-based ones for unrestricted texts. Unfortunately, the lack of large-scale tagged corpuses cause the supervised machine learning methods cannot be implemented to identify the geographical event related messages. In this paper, we propose an automatic method for recognizing micro-blogs that are related to geographical events based on the topic model and word vector. This method could achieve a satisfying identification result by increasing the corpus scale rapidly. Firstly, the topic model is capable to extract topics from documents. Thus, the web pages fetched by a search engine are grouped by the topics, and the corpus is obtained after combining the pages under the topics that are related to geographical events through judging their keywords of each topic. Secondly, the distributed representation word vector model is introduced to compensate the lack of context in the micro-blog, which is caused by its character count limit. These word vectors are integrated into the context semantic information from corpus training during the vector generation process. Thirdly, the correlation between the micro-blog message and the given geographical event is calculated and applied to determine whether this message contains the specified geographical event or not. In addition, some heuristic rules are used to correct the error correlations of very short messages. Experiments where the rainstorm is set as the targeting geographical event are conducted to validate the feasibility of this approach. The test conducted on Sina topic micro-blog shows that the F-1 of identification reaches 71.41% and is 10.79% higher than the traditional machine learning algorithm based on Support Vector Machine. Based on the premise that the precision loss is limited, the recall rate would rise with an increase in the corpus scale. The recognition precision could achieve 60% in a dataset containing five million micro-blog texts that simulating the actual data content and environment. These recognized event related micro-blogs could be used to extract detailed information elements in the future.
As the main form of representing the geographical information, geo-raster data contains abundant geographical knowledge. With the rapid development of earth observation technology, high-resolution geo-raster data has been widely applied to many research fields, such as landform, soil, environment and hydrology. With respect to this context, the contradiction between the saving and transferring of massive geo-raster data and a limited channel capacity has become increasingly prominent with regard to the intensive increase of data size. Data compression techniques provide the possibility to solve this problem. This paper studies the compression method of geo-raster data based on gridded DEMs for the purpose of realizing massive data’s online transmission. By analyzing the characteristics of geo-grid data, this paper proposes a new compression method named as the two-phase compression method, which combines the conversion compression and the coding compression based on the data fidelity and the real-time compression principle. Meanwhile, this paper establishes an assessment method of two-phase compression method from the perspectives of accuracy and efficiency. In order to test and verify the data fidelity and the compression performance of the two-phase compression method, this paper conducted several experiments on a 10-node server cluster under the Linux operating system by using different sizes of gridded DEMs. The experiment results showed that the proposed two-phase compression method has provided good data fidelity. It keeps the data accuracy on both of the numerical and the representation structure. At the same time, the compression ratio is generally above 50%, and the almost real-time decompression/compression efficiency also indicates that it has a good performance. The two-phase compression method can significantly reduce the time consumption of data transmission through network, and improve the efficiency of network transmission. In all, this two-phase compression method of geo-raster data presents a good universality, and it can provide a technical support to the application of geo-raster data, such as the high-performance geo-computation.
Existing approaches in finding the local co-location patterns have several shortcomings: (1) they depend on user predefining thresholds for proximity between the spatial feature instances and (2) the mining results miss the statistically significant explanation. In this paper, we proposed a new self-adaptive method for finding the local co-location patterns for spatial datasets containing continuous variables. The interestingness and indicator function and the proximity area that are defined based on the Voronoi diagrams are introduced. A proximity matrix is built to avoid user predefining thresholds for proximity. At last, the local Getis-Ord's
Ocean eddy is a dynamic phenomenon which continuously propagates and evolves. In oceanography, eddy variability is always a hot topic. In previous studies, the results from different aspects such as the literature study, observation and simple mathematical statistics, are relatively general and subjective. Eddies' zonal aggregated propagation and regionalization, which classified a large set of spatial objects into several homogeneous spatially contiguous regions, can help the discovery of zonal flow patterns. This study introduces the regionalization method into investigating the zonal moving characteristics of ocean eddies. And we present a quantitative and systematic method, which is a spatially constrained regionalization method based on the propagation of ocean eddies on spatial grids to explore and discover the characteristics and laws of eddies' propagation. Firstly, we construct a weighted network of eddies′ propagation through refactoring the eddy moving characteristics on the grid basis (each grid is a spatial sea unit), and then employs the spatially constrained regionalization method to divide the network of eddies' propagation into spatial continuous sub-regions. The adjacency-based regionalization method includes two steps: first, it constructs a hierarchy of clusters, which is a spatially contiguous tree, from the bottom up by iteratively merging the most connected nodes in the weighted network; and then, the spatially contiguous tree is partitioned from the top down, by finding the best edge to remove. A case study has been carried out on the ocean eddies identified in the South China Sea (SCS) for a period between 1992 and 2011. Three major regions are discovered by the proposed method: southeast of Vietnam (R1), east of Vietnam-Palawan (R2), and north of SCS (R3). These three regions are consistent with the discovered active propagation laws of ocean eddies in SCS: R1 covers the active band of ocean eddies in the deep sea basin of the south-western SCS; R2 well reflects the westward moving pattern of eddies in the central SCS; and R3 covers the sea areas along the northeast-southwest continental slope in the northern SCS. These findings show that the spatially constrained regionalization is efficient and quantitative, which is an appropriate method to study the moving patterns of ocean eddies. Also, we analyze the seasonal variation of eddies′ propagation within and across the three regions. The results demonstrate that there are different seasonal variations in the warm and cold eddies for the three regions. And the seasonal variations of eddies′ propagation in R1 and R3 is roughly similar: the warm eddies′ propagation in both R1 and R3 reach its peak in summer and fall but diminishes significantly in winter, whereas the cold eddies′ propagation, on the contrary, is weak in summer and fall, but increases gradually in winter and reaches its peak in spring. In R2, the warm eddies' propagation is most intensive in spring while the cold eddies' propagation is strong in summer and winter but weak in spring and fall. In summary, this paper provides a new perspective and approach to explore and study propagation characteristics and laws of ocean eddy by introducing the spatially constrained regionalization method creatively.
It is well known that glaciers in mountains are retreating widely in a warmer climate during recent decades on the Tibetan Plateau (TP). The runoffs from both glaciers and glacier lakes have been increased significantly, and the potential outbursts of glacier lakes have threaten the residence safety in China and the adjacent countries. However, most of the glaciers locate in very distant mountains. As glaciers are difficult to be investigated due to the huge investments and long travelling time of field survey, remote sensing monitoring has been the major approach adopted to understand the changes of glaciers nowadays. This paper has summarized several important items about current glacier studies, which includes: the development of remote sensing techniques on mountain glacier monitoring; the previous concluded results on glacier surface elevation changes in the major mountains on TP; and the problems and research trends of glacier studies based on the remote sensing techniques. Moreover, this paper reveals the glacier surface elevation changes on TP based on the ICESat/GLAS data. It shows that during 2003-2009 the glacier surface elevation on TP has changed by -0.24±0.03 m/a in average and yielding a mass change by -14.86±11.88 km3/a, whose melting water would run into rivers or lakes. The glacier change pattern on TP shows an obviously spatial-temporal heterogeneity, which decreases from the south and east TP toward the inland TP, and then it keeps decreasing toward the north and west TP.
Study on the segmentation method for high resolution remote sensing images is very important for the processing and application of remote sensing data. Image segmentation plays an important role in geographic object-based image analysis, and it is also very useful in GIS data management and remote sensing data compression. A new segmentation algorithm using optimized merging criteria is proposed in this paper. The proposed algorithm divides the merging process into two stages, including the local best merging and the global best merging. Hierarchical agglomerative clustering is used to implement the first stage to meet the main objective of increasing the running efficiency. The merging criterion in the first stage focuses on the regional geometric information to create the visually pleasing segments, and in addition, this criterion is constructed on the premise that the regions to be merged should be sufficiently similar in spectra. Thus, when designing the merging criterion of the local best merge, the spectral and geometric information are both taken into consideration. Moreover, Global Moran′s I is used to determine the ending condition for the first stage. After the local best merging, the region adjacency graph (RAG) is constructed to implement the global best merging, in which the spectral and edge information is taken into account. In this stage, the negative impact introduced by the regions′ scale is found throughout the experiments. Thus, the size information of each region is excluded from the merging criterion of the global best merging. In addition, a special binary search tree, which is called the red-black tree, is used in the implementation to rank the edges of RAG, so as to speed up the graph structure updating after a merging taking place. High resolution images acquired from OrbView3 are adopted to conduct the segmentation experiment, the results of which indicate that our algorithm can produce the satisfactory performance. The conclusions made in this paper may provide new insights for the studies on remote sensing image segmentation and the related researches.
Due to the variability and complexity of the topography in the Tibetan Plateau, the snow cover over most area of Tibetan Plateau is thin and revealing a high temporal and spatial heterogeneity. The passive microwave remote sensing greatly limits the precision of retrieved snow depth over the Tibetan Plateau, on account of its low spatial resolution and the uncertainty existed during snow depth retrieval. This paper attempts to reconstruct higher quality of snow depth data over the Tibetan Plateau through the fusion of multi-source remote sensing data, combining with a physic based snow model (SnowModel). This research mainly includes the following aspects: first of all, using the in-situ observed snow depth data and corresponding MODIS fractional snow cover data, the snow depletion curve of the study area is established. The MODIS fractional snow cover products (500 m) and passive microwave snow depth products (0.25°) are combined to produce the downscaled snow depth data (0.1°) using an empirical combination rule and the established snow depletion curve. Then, the downscaled snow depth data are assimilated into the SnowModel using the ensemble Kalman filter (EnKF) method. The accuracy of the downscaled snow depth data and the assimilated snow depth are analyzed through comparing them with the in situ observed snow depth data. The results show that there is an obvious depletion curving relationship between the snow depth and fractional snow cover area in the Tibetan Plateau. Using the root mean square error (RMSE) and correlation coefficients (R) as the evaluation standard, the assimilated snow depth is evaluated to be closer to the in-situ observed snow depth than the downscaled snow depth data.
In the field of remote sensing applications, vegetation indices have been used as an effective parameter for monitoring surface vegetation biophysical features such as the growth status and vegetation coverage. However, in the mountainous area, the influence of terrain causes the applicability of some vegetation indices being limited. It is conductive to improving the accuracy of vegetation indices by using topographic correction model in rugged terrain area. In this paper, 4 topographic correction models (Teillet-regression model, Minnaert model, C model and SCS + C model) were employed to correct several commonly used vegetation indices (SR, MSR, NDVI, SAVI, MSAVI and EVI) derived from the Landsat-5 TM data. This paper aimed at assessing the topographic correction results of these vegetation indices on different slope gradients, by taking Fanjing Mountain as the research area. The results indicated that topographic correction had different influences with respect to different vegetation indices. Topographic correction was more effective on non-band-ratio vegetation indices (SAVI, MSAVI and EVI) than band-ratio vegetation index (SR, MSR and NDVI), because the band ratio could reduce the effect of topography to some extent. Furthermore, the effects of topographic correction on the vegetation index were also different with respect to different slopes. While the slope increases, the topographic effects became more significant and topographic correction acted more effectively. On the gentler slopes, the non-band-ratio vegetation indices, rather than the band-ratio vegetation indices, need to be corrected using topographic correction models. On the steep slopes, we recommended that both of the indices should be corrected, while the non-band-ratio vegetation indices may be overcorrected. In addition, the precision of the linear regression equation of the non-band-ratio vegetation indices and forest aboveground biomass was improved after the application of topographic correction. Above all, before using the non-band-ratio vegetation indices for quantitatively retrieving the vegetation parameters in rugged terrain area, topographic correction is recommended to be conducted.
Law enforcement and supervision of land resources are important components in the land resources management systems. Their current way of working is mainly manipulated by the traditional manual operation, which is not efficient in performance. With the launch of high-spatial-resolution remote sensing satellites, the application of details contained within these received images makes it possible to accurately and rapidly enforce the law and supervise the land resources with the adoption of remote sensing technology. Therefore, in order to analyze the actual requirements of the applications and the latest technological developments, this paper intends to extract the illegal construction land information using the high-spatial-resolution remote sensing images, and we choose the Huangyan district to be the demonstration area to carry out the county-level law enforcement and to implement proper supervision to land resources. We have extracted the suspected illegal build-up construction sites using the change detection and urban extraction technologies. The experiment achieves a good result and demonstrates the significant potentials of domestic high-spatial-resolution satellite in the practical applications of law enforcement and supervision of land resources.
As an important indicator in measuring the economic development level of a region, GDP spatialization is of great significance to study the socio-economic heterogeneity. The ancillary spatial density data selection is the key technique in controlling the GDP spatialization′s accuracy. In this paper, the prefectural GDP statistics is distributed to grid cells according to the spatial distribution information of GDP such as the population density (LandScan, AsiaPop) and night light data in Beijing-Tianjin-Hebei. Moreover, the absolute errors and relative errors of the GDP disaggregation at county-level are both calculated in order to compare the errors among the three different ancillary data as mentioned above. These results can provide a reasonable reference to ancillary spatial density data selection in GDP disaggregation. The results show that, the spatial distributions of the three types of ancillary spatial density data for GDP have revealed their own advantages and disadvantages. Comparing with both of the night light and the LandScan data, the AsiaPop simulation generally has the smallest error, especially in the suburban districts and rural areas of Beijing where the GDP tends to be overestimated, while the GDP is often underestimated in the economically developed city centers. For the LandScan simulation, six counties have presented a relative error of more than 200%, as the LandScan data are concentrated in Beijing and Tianjin, while the suburban districts and counties have also been overestimated. The AsiaPop simulation has only three counties (which locate in Tianjin) presenting a relative error being more than 200%. Because of the spatial heterogeneity of the economic activities, the GDP disaggregation error will increase with respect to the refinement of the administrative units, therefore, using the single-generation data to reasonably reflect the spatial distribution of economic activities is difficult, we need to take advantage of the distribution data such as the night light, roads, housing distribution and cell phone signals to improve the GDP disaggregation′s accuracy in future, and to reflects the GDP distribution characteristics in a more detailed manner. High-quality exposure data not only provide the basic data for the study of spatial analysis of natural disaster risk, but also provide a reference for other multidisciplinary research fields; meanwhile, the comprehensive application of using both the multi-source remote sensing data and the statistics data is the trend for socio-economic data spatialization.
The Heihe River Basin is located in the northwest arid region of China. Water shortage is the primary restrictive factor for oasis agriculture development and ecological environment protection in this region. Since the water consumed by the oasis agriculture in the middle reaches occupies about 80% of the total water resources in this basin, how to efficiently use the agricultural water resources is the key factor affecting the regional development. In the exploration of agricultural water saving measures in the arid region, most scholars advocate to improve the water use efficiency (WUE) through water-saving techniques. However, there are few quantitative studies focusing on the water saving effect of planting structure adjustment. This study takes the adjustment of agricultural planting structure as a breakthrough point to achieve the goal of agricultural water saving, considering that different crops have different water demands. We aim to explore the characteristics of the water demands of the main crops in the middle reaches of Heihe River and examine the different levels of WUE for those crops. In this paper, with the help of ArcGIS technique and based on the evapotranspiration (ET) data of Heihe River Basin and crop spatial distribution data, we extracted the ET information of the main crops during their growing seasons in 2012, which was represented as the water requirement of crop. Then, the crop ET data and crop yield data were used to calculate the WUE of four main crops. The results show that: (1) maize has the largest ET during its growing seasons, followed by wheat, rapeseed and barley; (2) considering the case of precipitation recharge, the water demands of barley and rapeseed may largely depend on rainfall, while maize and wheat require irrigation instead, and maize needs more water than wheat; (3) the WUE of crops ranking from high to low is barley, rapeseed, wheat and maize. This study argues that the appropriate increase in the size of wheat planting is helpful to improve the efficiency of agricultural water in the middle reaches of Heihe River.
As the longest tributary on the right bank of upper Yangtze River, while also occupying the largest basin area in Guizhou Province, Wujiang River is of great importance to the economic and ecological environment in Yangtze Basin and Southwest China. However, the upper reaches of Wujiang River are suffering from long-term soil erosion and land degradation, which has threatened the local safety and development. This paper, based on the forest resource inventory data of Bijie prefecture in Guizhou Province in 2010, aims to estimate the water conservation of forest ecosystem in the upper reaches of Wujiang River and to analyze its spatial variation. The relationship between the unit water conversation of forest ecosystem, which is regarded as the Forest Water Conservation Capacity (FWCC) in this paper, and elevation, slope and land degradation types was deeply explored. The integrated storage capacity method and the linear regression was employed. The results show that: (1) In 2010, the water conservation of forest ecosystem in the study area was 563.05 million cubic meters in total, yielding a water conservation of 774.73 t/hm2 per unit area. FWCC presented a pattern of gradually decreasing from northeast to southwest in the east region, and an uneven pattern in the west region. (2) A significant
Located in the north of Central Asia, Kazakhstan is among regions having the highest fragility of eco-environment in the world. With the implementation of “the Silk Road Economic Belt and the 21st-Century Maritime Silk Road” strategy advocated by Chinese government since 2013, more attention is being paid to researches focusing on the eco-environment of this country. Utilizing the land use data interpreted by TM images with a spatial resolution of 30 meters, the Ecological Quality Indices (EQIs) of Kazakhstan in 2000 and 2010 were evaluated by considering its biological richness, vegetation coverage, water network denseness and land deterioration. The evaluation results indicated that the EQI values varied among different prefectures. The eastern states showed a relative higher level of eco-environment quality than the western ones, and an overall declining pattern of the quality level was revealed from 2000 to 2010. In order to explore the driving forces of the spatial variation of EI among prefectures, natural factors such as NDVI, Modified Surface Water Capacity Index (MSWCI) and Land Surface Temperature (LST) were retrieved from the MODIS data, and subsequently input into the linear regression model as the independent variable together with the social economic index such as population density and real GDP per capita by taking EQI as the dependent variable. Regression results indicated that EIs in Kazakhstan were mainly influenced by the natural factors, especially by NDVI which represented the vegetation coverage. Thus, the key point is to raise the vegetation coverage in order to improve the eco-environment quality in Kazakhstan.
