Liu Yaolin

Rural land use spatial allocation in the semiarid loess hilly area in China: Using a Particle Swarm Optimization model equipped with multi-objective optimization techniques

Semiarid loess hilly areas in China are enduring a series of environmental conflicts between urban expansion, cultivated land conservation, soil erosion and water shortage, and require land use allocation to reconcile these environmental conflicts. We argue that the optimized spatial allocation of rural land use can be achieved by a Particle Swarm Optimization (PSO) model in conjunction with multi-objective optimization techniques. Our study focuses on Yuzhong County of Gangsu Province in China, a typical catchment on the Loess Plateau, and proposes a land use spatial optimization model. The model maximizes land use suitability and spatial compactness based on a variety of constraints, e.g. optimal land use structure and restrictive areas, and employs an improved PSO algorithm equipped with a determinant initialization method and a dynamic weighted aggregation (DWA) method to obtain the optimized land use spatial pattern. The results suggest that (1) approximately 4% of land use should be reallocated and these changes would alleviate the environmental conflicts in the study area; (2) the major reshuffling is slope farmland and newly added construction and cultivated land, whereas the unchanged areas are largely forests and basic farmland; and (3) the PSO is capable of optimizing rural land use allocation, and the determinant initialization method and DWA can improve the performance of the PSO.

Model of Land Suitability Evaluation Based on Computational Intelligence

A novel model of land suitability evaluation is built based on computational intelligence (CI). A fuzzy neural network (FNN) is constructed by the integration of fuzzy logic and artificial neural network (ANN). The structure and process of this network is clear. Fuzzy rules (knowledge) are expressed in the model explicitly, and can be self-adjusted by learning from samples. Genetic algorithm (GA) is employed as the learning algorithm to train the network, and makes the training of the model efficient. This model is a self-learning and self-adaptive system with a rule set revised by training.

Frameworks for generalization constraints and operations based on object-oriented data structure in database generalization

The constraints and the operations play an important role in database generalization. They guide and govern database generalization. The constraints are translation of the required conditions that should take into account not only the objects and relationships among objects but also spatial data schema (classification and aggregation hierarchy) associated with the final existing database. The operations perform the actions of generalization in support of data reduction in the database. The constraints in database generalization are still lack of research. There is still the lack of frameworks to express the constraints and the operation on the basis of object-oriented data structure in database generalization. This paper focuses on the frameworks for generalization operations and constraints on the basis of object-oriented data structure in database generalization. The constraints as the attributes of the object and the operations as the methods of the object can be encapsulated in classes. They have the inheritance and polymorphism property. So the framework of the constraints and the operations which are based on object-oriented data structure can be easily understood and implemented. The constraint and the operations based on object-oriented database are proposed based on object-oriented database. The frameworks for generalization operations, constraints and relations among objects based on object-oriented data structure in database generalization are designed. The categorical database generalization is concentrated on in this paper.

A study on estimation of the amount of soil erosion in small watershed based on GIS: a case study in the Three Gorge Area of China

Soil erosion is a serious environmental and production problem in the Three Gorge Areas (TGA) of China. To decrease the risk on environmental impacts, there is an increasing demand for sound, and readily applicable techniques for soil conservation planning in this area. The objectives of the study were to develop and validate a soil erosion-predicting model based on the revised Universal Soil Loss Equation (RUSLE) in a geographic information systems (GIS) environment. The use of GIS to develop conservation-oriented watershed management strategies in the Taipingxi watershed is presented. Data used for the RUSLE were either determined or taken from published literature pertaining to the Taipingxi watershed. The study showed that the serious eroded area (sediment is higher than 50 t/ha) is 10.62%, but contributes 61.55% sediments of all watershed, while no or slightly eroded area (sediment is lower than 10t/ha) is 62.18%, only contributes 3.49% sediments in the watershed. In the watershed, the annual average soil loss rate from relatively flat agricultural land was approximately 21 t/ha, whereas 48 t/ha was found on the cultivated sloping lands, which constitutes a large proportion of soil loss. The results of the study imply the capability of the RUSLE in the GIS environment to predict soil loss quantitatively and spatially. And they also indicate that the RUSLE-GIS model is a useful tool for resource management and soil conservation planning. This technology is readily transferable and accessible to other land managers and agronomists in the TGA.

Spatial object aggregation based on data structure, local triangulation and hierarchical analyzing method

This paper focuses on the methods and process of spatial aggregation based on semantic and geometric characteristics of spatial objects and relations among the objects with the help of spatial data structure (Formal Data Structure), the Local Constrained Delaunay Triangulations and semantic hierarchy. The adjacent relation among connected objects and unconnected objects has been studied through constrained triangle as elementary processing unit in aggregation operation. The hierarchical semantic analytical matrix is given for analyzing the similarity between objects types and between objects. Several different cases of aggregation have been presented in this paper.

Categorical database generalization

This paper focuses on the issues of categorical database generalization and emphasizes the roles of supporting data model, integrated data model, spatial analysis and semantic analysis in database generalization. The framework contents of categorical database generalization transformation are defined. This paper presents an integrated spatial supporting data structure, a semantic supporting model and similarity model for the categorical database generalization. The concept of transformation unit is proposed in generalization.

Multi-scale Representation of Building Feature in Urban GIS

This paper aims at multiscale representation of urban GIS, presenting a model to dynamically generalize the building on the basis of Delaunay triangulation model. Considering the constraints of position accuracy, statistical area balance and orthogonal characteristics in building cluster generalization, this paper gives a progressive algorithm of building cluster aggregation, including conflict detection (where), object (who) displacement, and geometrical combination operation (how). The algorithm has been realized in an interactive generalization system and some experiment illustrations are provided.

Spatial relation resolution and spatial relation abstraction

This paper attempts to regard spatial relation transformation as an important process in map generalization. The spatial relation generalization can be divided into the components of abstraction: topology, disttance and orientation. The concept ‘spatial relation resolution’ is introduced to describe the constraints of relative spatial relation. On the basis of nine intersection models, the cardinal direction models and the iso-distance-relation models, this paper gives three sorts of relation resolution representations for topological, distance and orientation relation, respectively. Two mapping implementations in map generalization is discussed.

Design of the urban land grading information system

This paper discusses the urban land grading system based on the self-developed GIS software. The system establishes the system template, the relevant data structure and the empirical formulas of the affecting factor and economic data of land use. Besides, the system can calculate the service radius, sum and land grade automatically. Furthermore, it can do statistics and query on various land information and show the result of land grading with spatial and attribute data. This paper illustrates the result by giving an example of the commercial land grading of the urban area of Wuhan City by the urban land grading system.

Discriminant analysis on land grading

This paper proposes the discriminant analysis on land grading after analyzing the common methods and discussing the Fishers discriminant in detail. Actually this method deduces the dimension from multi to single, thus it makes the feature vectors inn-dimension change to a scalar, and use this scalar to classify samples. This paper illustrates the result by giving an example of the residential land grading by the discriminant analysis.