Key Engineering Materials Vols. 467-469

Abstract: Graph cuts as an increasingly important tool for solving a number of energy minimization problems in computer vision and other fields, meanwhile beamlet transform as time-frequency and multiresolution analysis tool is often used in the domain of image processing, especially for image fusion. By analyzing the characters of DSA medical image, this paper proposes a novel DSA image fusion method which is combining beamlet transform and graph cuts theory. Firstly, the image was decomposed by beamlet transform to obtain the different subbands coefficients. Then an energy function based on graph cuts theory was constructed to adjust the weight of these coefficients to obtain an optimum fusion object. At last, an inverse of the beamlet transform reconstruct a synthesized DSA image which could contain more integrated accurate detail information of blood vessels. By contrast, the efficiency of our method is better than other traditional fusion methods.

Abstract: A dynamic method based on improved algorithm BP neural network for damage identification of composite materials was proposed. By using wavelet series, the features of signals were extracted and input to improved algorithm BP neural network for training the network and identifying the damages. Finally, the experiment results show that this proposed method can exactly identify the faults of composite materials.

Abstract: Pipelined operator graph (POG) scheduling is an important problem in the area of parallel query optimization. A POG is a graph with vertices representing query operators that can run in parallel and edges representing communication between adjacent operators. The problem is to assign operators to processors so as to minimize the maximum processor load. We present a 2-approximation algorithm for the case where the operator graph has bounded treewidth.

Abstract: This paper presents the work done towards searching a collision-free path for mobile crane based on C-space in the complex 3D working environment. The crane is simplified into three degrees of freedom (DOFs) robot, each of which is represented as an axis of configuration space (C-space). In this paper, we propose an improved ant colony approach for crane path planning, which takes into full account of not only the factor of the shortest path but also the factor of safety. In this approach, we employ more complete heuristic information, introduce adaptive pheromone volatilization coefficient and pheromone penalty factors, and prevent ants from falling into trap and the stagnation. The reasonability and practicability of the proposed approach for automated path planning is verified by comparing the performances of the present approaches in the practice case, and the comparison results show that the algorithm can gain a relatively optimal solution in short time and have a great value of engineering application.

Abstract: For the model of induction motors(IMs) in field-oriented coordinates, a novel design method of controller is proposed, which combining block adaptive backstepping method with neural networks, introducing PD-type feedback, and making use of the diagonal feature of the unknown control matrix and the boundedness of its derivative. The control law and parameter updating law are derived using Lyapunov theory, which guarantees the stability of the whole system. The proposed approach can track the rotor speed and flux reference signals under parameter uncertainties. Simulation results show the effectiveness of the proposed approach.

Abstract: Performance evaluation of enterprise is a summary of corporate management and it is an important part of measures to improve efficiency. So, It is concerned by the business community, academia and government departments. This paper attempts to use fuzzy comprehensive evaluation method to do performance evaluation of enterprises. Based on theoretical analysis, the paper study the case. The conclusion is reasonable. So, the method is valuable and scientific.

Abstract: This paper analyzed the existing association rules update algorithm IUA, found out that when the decision makers gave priority attention to the situation of maximum frequent itemsets, this algorithm cannot lower the cost of the database traversal to quickly access to the largest number of frequent itemsets. For the lack of the algorithm, an algorithm which is based on reverse search approach to update association rules is presented. The updating algorithm based on reverse search first generated all frequent itemsets of new itemsets. Then, it spliced the new largest frequent itemsets and original largest frequent itemsets for trimming, get the updated maximal frequent itemsets. This algorithm not only reduces the traversal times in the process of association rules updating, but also realized the priority access to the largest operation of frequent itemsets.

Abstract: The dynamics and uncertainty of the business and the market makes difficult to coordinate the activities of a supply chain. Therefore, it is important to review systematically and to take into account the variability in the planning formulation in order to manage a supply chain network efficiently. A novel stochastic multi-period design and planning MILP model of a multiechelon supply chain network is used as a predictive model in this work. Model predictive control is presented as a way to manage supply chain in the presence of uncertainty by incorporating unforeseen events into the planning process. Illustrative example shows control strategy based on model predictive control framework is effective in the supply chain network design and planning.

Abstract: Information is the soul of the logistics, and the powerful logistics network is a prerequisite for the Fourth Party Logistics. Information network can control all aspects of the logistics supply chain, integrate the national logistics resources effectively, enhance the efficiency and reduce logistics cost. This paper proposes a logistics platform based on WebGIS/3G/GPS, which can share the attribute data and map data in high degree and realize distribute storage. It not only can change the dull attribute data into vivid graphics, but also makes the management and maintenance for massive data convenient and efficient, and control the logistics process accurately and timely. The implementation of the platform will maximize the advantages of the Fourth Logistics, strengthen control over the logistics process.

Abstract: After the occurrence of significant natural disaster, the resulting damaged roads interrupt the rapid emergency response for disaster, and therefore, the disaster relief department is desperate for the destruction condition of roads in the devastated region, which can help make relief decisions and deploy rescue actions. In view of the practical needs of the disaster relief department and the objective fact that at present there is not any special, high automatic damage detection system of roads, we develop Road Damage Detection and Evaluation System (RODDES). Using the basic road data in GIS (Geographical information system) as the prior knowledge, the system extracts the pre-disaster and post-disaster roads from post-disaster remotely sensed imageries, and then detects the damaged regions and evaluates the destruction condition. This paper emphasizes the overall design of the system and the submodule design and their functions. The system is applied in detecting and evaluating the damaged roads in Wenchuan County, China and the experiment results show that nearly all producer’s and user’s accuracies of the road extractions and damage detections are above 75%, and it accurately evaluates the destruction condition of roads.