Search

We propose an intelligent tool for terrestrial transportation design based on the underlying concept of adaptive meshing. Vehicle routes are deformable visual patterns which cover the geographic area adapting shapes to distributed demands or specific places. In this way the system allows dynamic interaction and continuous visual feedback for the designer constructing and evaluating a transportation bus network. To assist the designer, an optimization framework merging neural networks and evolutionary algorithms permits one to position services and create routes among them automatically. The transport system optimizer built on a geographic information system is illustrated by application to a real life case of clustering and routing for the transportation of the 780 employees of an enterprise. [ABSTRACT FROM AUTHOR]

This work focuses on the dynamic Pickup and Delivery Problem with Time Windows (PDPTW). The transport requests should be performed using the available fleet of vehicles. The vehicles move between the nodes of a road network. The aim of this work is to propose a model which allows, during a transport plan creation, to take into account predictable events. Particularly, we consider the frequency of requests at any node in the road network and the construction of vehicle routes that will allow new requests to be inserted without any significant route modification. Therefore, we construct routes that pass near the nodes where transport requests are most frequently generated.

In this paper, we focus on the dimensioning process of cellular networks that addresses the evaluation of equipment global costs to cover a city. To deal with frequency assignment, that constitutes the most critical resource in mobile systems, the network is usually modeled as a pattern of regular hexagonal cells. Each cell represents the area covered by the signal of a transmitter or base station (BS). Our work emphasizes on the design of irregular hexagonal cells in an adaptive way. Hexagons transform themselves and adapt their shapes according to a traffic density map and to geometrical constraints. This process, called adaptive meshing (AM), may be seen as a solution to minimize the required number of BS to cover a region and to propose a basis for transmitter positioning. The solution we present to the mesh generation problem for mobile network dimensioning is based on the use of an evolutionary algorithm. This algorithm, called hybrid island evolutionary strategy (HIES), performs distributed computation. It allows the user to tackle problem instances with large traffic density map requiring several hundreds of cells. HIES combines local search fast computation on individuals, incorporated into a global island-like strategy. Experiments are done on one real case representing the mobile traffic load of the second French city of Lyon and on several other traffic maps from urban fictive data sets. [ABSTRACT FROM AUTHOR]

With the development of computer capabilities, memories and network abilities, we need more efficient and robust algorithms to manage databases and to store and retrieve the relevant information for the user. The aim of this work is to automate the construction of a neural network Information Retrieval System (IRS) adapted to a medical image database. The user builds queries and the system must retrieve the relevant documents or images. Queries are groups of keywords or items associated with relevant images. In our approach, the set of queries and the binary relevance judgments on the documents constitute complex learning data associations. There are two phases in the automatic construction of the IRS. The indexing phase builds the learning data base and then a specific learning algorithm builds the neural network. For the system to be able to immediately learn these complex data, we have developed a new specific algorithm. It allows a perfect learning of a binary logical table in a stepwise fashion without forgetting the previously learnt logical combinations. Furthermore, this algorithm works very quickly and leads to a parallel implementation for large databases. [ABSTRACT FROM AUTHOR]

Automatically finding correspondences between object features in images is of main interest for several applications, as object detection and tracking, identification, registration, and many derived tasks. In this paper, we address feature correspondence within the general framework of graph matching optimization and with the principal aim to contribute. We proposed two optimized algorithms: first-order and second-order for graph matching. On the one hand, a first-order normalized cross-correlation (NCC) based graph matching algorithm using entropy and response through Marr wavelets within the scale-interaction method is proposed. First, we proposed a new automatic feature detection processing by using Marr wavelets within the scale-interaction method. Second, feature extraction is executed under the mesh division strategy and entropy algorithm, accompanied by the assessment of the distribution criterion. Image matching is achieved by the nearest neighbor search with normalized cross-correlation similarity measurement to perform coarse matching on feature points set. As to the matching points filtering part, the Random Sample Consensus Algorithm (RANSAC) removes outliers correspondences. One the other hand, a second-order NCC based graph matching algorithm is presented. This algorithm is an integer quadratic programming (IQP) graph matching problem, which is implemented in Matlab. It allows developing and comparing many algorithms based on a common evaluation platform, sharing input data, and a customizable affinity matrix and matching list of candidate solution pairs as input data. Experimental results demonstrate the improvements of these algorithms concerning matching recall and accuracy compared with other algorithms. [ABSTRACT FROM AUTHOR]

This paper proposes a parallel computation model for the self-organizing map (SOM) neural network applied to Euclidean traveling salesman problems (TSP). This model is intended for implementation on the graphics processing unit (GPU) platform. The Euclidean plane is partitioned into an appropriate number of cellular units, called cells, and each cell is responsible of a certain part of the data and network. Compared to existing GPU implementations of optimization metaheuristics, which are often based on data duplication or mixed sequential/parallel solving, the advantage of the proposed model is that it is decentralized and based on data decomposition. Designed for handling large-scale problems in a massively parallel way, the required computing resources grow linearly along the problem size. Experiments are conducted on 52 publicly available Euclidean TSP instances with up to 85,900 cities for the largest TSPLIB instance and 71,009 cities for the largest National TSP instance. Experimental results show that our GPU implementations of the proposed model run significantly faster than the currently best-performing neural network approaches, to obtain results of similar quality. [ABSTRACT FROM AUTHOR]