The objective of this study is to develop a unique modeling approach for fast assessment of massive soil erosion by water at a regional scale in the Loess Plateau, China. This approach relies on an understanding of bo...The objective of this study is to develop a unique modeling approach for fast assessment of massive soil erosion by water at a regional scale in the Loess Plateau, China. This approach relies on an understanding of both regional patterns of soil loss and its impact factors in the plateau area. Based on the regional characteristics of precipitation, vegetation and land form, and with the use of Landsat TM and ground investigation data, the entire Loess Plateau was first divided into 3 380 Fundamental Assessment Units （FAUs） to adapt to this regional modeling and fast assessment. A set of easily available parameters reflecting relevant water erosion factors at a regional scale was then developed, in which dynamic and static factors were discriminated. Arclnfo GIS was used to integrate all essential data into a central database. A resulting mathematical model was established to link the sediment yields and the selected variables on the basis of FAUs through overlay in GIS and multiple regression analyses. The sensitivity analyses and validation results show that this approach works effectively in assessing large area soil erosion, and also helps to understand the regional associations of erosion and its impact factors, and thus might significantly contribute to planning and policymaking for a large area erosion control in the Loess Plateau.展开更多
Support vector machine(SVM) has shown great potential in pattern recognition and regressive estima-tion.Due to the industrial development demands,such as the fermentation process modeling,improving the training perfor...Support vector machine(SVM) has shown great potential in pattern recognition and regressive estima-tion.Due to the industrial development demands,such as the fermentation process modeling,improving the training performance on increasingly large sample sets is an important problem.However,solving a large optimization problem is computationally intensive and memory intensive.In this paper,a geometric interpretation of SVM re-gression(SVR) is derived,and μ-SVM is extended for both L1-norm and L2-norm penalty SVR.Further,Gilbert al-gorithm,a well-known geometric algorithm,is modified to solve SVR problems.Theoretical analysis indicates that the presented SVR training geometric algorithms have the same convergence and almost identical cost of computa-tion as their corresponding algorithms for SVM classification.Experimental results show that the geometric meth-ods are more efficient than conventional methods using quadratic programming and require much less memory.展开更多
基金Under the auspices of Northeast Normal University Sci-tech Innovation Incubation Program(No.NENU-STC08017)European Commission FP7 Project―PRACTICE(No.ENVI-2008-226818)
文摘The objective of this study is to develop a unique modeling approach for fast assessment of massive soil erosion by water at a regional scale in the Loess Plateau, China. This approach relies on an understanding of both regional patterns of soil loss and its impact factors in the plateau area. Based on the regional characteristics of precipitation, vegetation and land form, and with the use of Landsat TM and ground investigation data, the entire Loess Plateau was first divided into 3 380 Fundamental Assessment Units （FAUs） to adapt to this regional modeling and fast assessment. A set of easily available parameters reflecting relevant water erosion factors at a regional scale was then developed, in which dynamic and static factors were discriminated. Arclnfo GIS was used to integrate all essential data into a central database. A resulting mathematical model was established to link the sediment yields and the selected variables on the basis of FAUs through overlay in GIS and multiple regression analyses. The sensitivity analyses and validation results show that this approach works effectively in assessing large area soil erosion, and also helps to understand the regional associations of erosion and its impact factors, and thus might significantly contribute to planning and policymaking for a large area erosion control in the Loess Plateau.
基金Supported by the National Natural Science Foundation of China (20476007,20676013)
文摘Support vector machine(SVM) has shown great potential in pattern recognition and regressive estima-tion.Due to the industrial development demands,such as the fermentation process modeling,improving the training performance on increasingly large sample sets is an important problem.However,solving a large optimization problem is computationally intensive and memory intensive.In this paper,a geometric interpretation of SVM re-gression(SVR) is derived,and μ-SVM is extended for both L1-norm and L2-norm penalty SVR.Further,Gilbert al-gorithm,a well-known geometric algorithm,is modified to solve SVR problems.Theoretical analysis indicates that the presented SVR training geometric algorithms have the same convergence and almost identical cost of computa-tion as their corresponding algorithms for SVM classification.Experimental results show that the geometric meth-ods are more efficient than conventional methods using quadratic programming and require much less memory.
