Intelligent Transportation System (ITS) is a system that builds a safe, effective and integrated transportation environment based on advanced technologies. Road signs detection and recognition is an important part of ITS, which offer ways to collect the real time traffic data for processing at a central facility.This project is to implement a road sign recognition model based on AI and image analysis technologies, which applies a machine learning method, Support Vector Machines, to recognize road signs. We focus on recognizing seven categories of road sign shapes and five categories of speed limit signs. Two kinds of features, binary image and Zernike moments, are used for representing the data to the SVM for training and test. We compared and analyzed the performances of SVM recognition model using different features and different kernels. Moreover, the performances using different recognition models, SVM and Fuzzy ARTMAP, are observed.

Walking machines may not be more common or faster than the transportations with wheels. It can¡¦t be ignored in the occasions of unknown terrain. This paper is going to discuss about how a walking machine get faster and be static stable. When the quadrupeds walk toward, the wide won¡¦t be changed. So that, longitudinal stability margin can take the place of stability margin to simplify gait problems. Meanwhile we can get the optimal gait. In the past researches, there is only one kind of walking type will be discussed in one time. This is because there are not so many relationships between different kinds of movement. If we take one step ahead to discuss the optimal gait, it will be more difficult. If there was a way to get into optimal gait from random initial position, we can connect one movement with the other. The velocity was constrained while the quadruped modal has had been made since 1968 by McGhee. We will try to change the working area to approve the performance. As to the researches of multi-legs walking machine, most of them talk about quadrupeds and hexapods. it will be less if the more legs we are talked about. To maintain stable tread, a walking machine request four legs at least. We can regard a quadruped as a unit, and divide a multi-leg working machine in to many quadrupeds. By using the method of quadruped analysis, we can simplify multi-legs gait algorithm problems.

This thesis proposes an approach to segmenting and identifying mixed-language speech. Automatic LID can be divided into four steps, feature extraction, segmentation, segment clustering, and re-labeling. In feature extraction, we compare the group delay feature (GDF) with MFCC feature. Unlike the traditional feature from Fourier trans-form magnitude, GDF uses the phase spectrum. In segmentation, we compare delta Bayesian information criterion (delta-BIC) with support vector machines (SVMs). A delta-BIC is applied to segment the input speech utterance into a sequence of lan-guage-dependent segments using acoustic features. The segments are clustered using the K-means algorithm. Finally, re-labeling is used to determine the language of the clusters. SVMs proceed to segment and identify automatically after model training. Considering the effect of the accent issue, we use the corpus English Across Taiwan (EAT) to perform our system. The experimental results show that the system can reach 78.13% in the frame hit rate under the baseline 57.77%.

Texte remanié de: Proefschrift--Geschiedenis--Universiteit van Amsterdam, 2007. / Bibliogr. p. 345-358. Index. Résumé en anglais.

Thèse de doctorat : Sciences. Mécanique : Ecole centrale de Lyon : 2003. / 162 réf.

Thèse de doctorat : Sciences. Mécanique : Ecole centrale de Lyon : 2003. / Titre provenant de l'écran-titre. Bibliogr. 210 réf.

Thèse de doctorat : Sciences, Mécanique : Ecole centrale de Lyon : 2003. / Titre provenant de l'écran-titre. 162 réf.

Thèse de doctorat : Sciences. Mécanique : Ecole centrale de Lyon : 2003. / 210 réf.

Reproduction de : Thèse de doctorat : Génie électrique : Toulouse, INPT : 2004. / Titre provenant de l'écran-titre. Bibliogr. 74 réf.

Thèse de doctorat : Génie électrique : Lille 1 : 2002. Thèse de doctorat : Génie électrique : Université Politehnica de Bucarest : 2002. / Thèse soutenue en cotutelle. N° d'ordre (Lille) : 3107. Résumé en français, roumain et anglais. Bibliogr. p. 89-93.