Search results for "pattern recognition"
showing 10 items of 2301 documents
Health Indicator for Low-Speed Axial Bearings Using Variational Autoencoders
2020
This paper proposes a method for calculating a health indicator (HI) for low-speed axial rolling element bearing (REB) health assessment by utilizing the latent representation obtained by variational inference using Variational Autoencoders (VAEs), trained on each speed reference in the dataset. Further, versatility is added by conditioning on the speed, extending the VAE to a conditional VAE (CVAE), thereby incorporating all speeds in a single model. Within the framework, the coefficients of autoregressive (AR) models are used as features. The dimensionality reduction inherent in the proposed method lowers the need of expert knowledge to design good condition indicators. Moreover, the sugg…
Power losses in power-split CVTs: A fast black-box approximate method
2018
Abstract This paper addresses the mechanical losses of planetary transmissions, with particular attention to power-split CVTs in their hybrid electric versions. It provides unified layout-independent analytical relationships, which can be used for both analysis, design and control purposes, and a simplified approach; the latter overcomes the necessity to segment the operating range of the power-split CVT in order to keep its loss model physically consistent. An example of application to a real hybrid electric PS-CVT is performed to show the simplicity, accuracy and generality of the proposed method.
Central catadioptric image processing with geodesic metric
2011
International audience; Because of the distortions produced by the insertion of a mirror, catadioptric images cannot be processed similarly to classical perspective images. Now, although the equivalence between such images and spherical images is well known, the use of spherical harmonic analysis often leads to image processing methods which are more difficult to implement. In this paper, we propose to define catadioptric image processing from the geodesic metric on the unitary sphere. We show that this definition allows to adapt very simply classical image processing methods. We focus more particularly on image gradient estimation, interest point detection, and matching. More generally, th…
A Geometrical Approach for Vision Based Attitude and Altitude Estimation for UAVs in Dark Environments
2012
International audience; This paper presents a single camera and laser system dedicated to the realtime estimation of attitude and altitude for unmanned aerial vehicles (UAV) under low illumination conditions to dark environments. The fisheye camera allows to cover a large field of view (FOV). The approach, close to structured light systems, uses the geometrical information obtained by the projection of a laser circle onto the ground plane and perceived by the camera. We propose some experiments based on simulated data and real sequences. The results show good agreement with the ground truth values from the commercial sensors in terms of its accuracy and correctness. The results also prove i…
K-nearest neighbor driving active contours to delineate biological tumor volumes
2019
Abstract An algorithm for tumor delineation in positron emission tomography (PET) is presented. Segmentation is achieved by a local active contour algorithm, integrated and optimized with the k-nearest neighbor (KNN) classification method, which takes advantage of the stratified k-fold cross-validation strategy. The proposed approach is evaluated considering the delineation of cancers located in different body districts (i.e. brain, head and neck, and lung), and considering different PET radioactive tracers. Data are pre-processed in order to be expressed in terms of standardized uptake value, the most widely used PET quantification index. The algorithm uses an initial, operator selected re…
Accurate keyframe selection and keypoint tracking for robust visual odometry
2016
This paper presents a novel stereo visual odometry (VO) framework based on structure from motion, where a robust keypoint tracking and matching is combined with an effective keyframe selection strategy. In order to track and find correct feature correspondences a robust loop chain matching scheme on two consecutive stereo pairs is introduced. Keyframe selection is based on the proportion of features with high temporal disparity. This criterion relies on the observation that the error in the pose estimation propagates from the uncertainty of 3D points—higher for distant points, that have low 2D motion. Comparative results based on three VO datasets show that the proposed solution is remarkab…
A macroscale FEM-based approach for selective laser sintering of thermoplastics
2017
A numerical approach to model the selective laser sintering (SLS) of polypropylene is proposed. A 3D thermal model was developed and thus enables the prediction of the temperature fields and the extension of the sintered area in the powder bed taking into account the phase change during multiple laser passes. Powderâliquid, liquidâsolid and solidâliquid phase changes were modelled during the SLS and the subsequent cooling processes. Then, a 3D thermomechanically coupled model was set up based on the temperature results of the thermal model in order to predict the distortion of the produced parts after cooling down. Different pre-heating temperatures were considered, highlighting their…
A numerical model for Wire integrity prediction in Friction Stir Extrusion of magnesium alloys
2017
Abstract A numerical model for the prediction of the wire quality produced by the novel direct machining chip recycling technique known as Friction Stir Extrusion (FSE) is presented. Wire microstructure and wire integrity have been predicted by embedding in the code the equations enabling the calculation of the Zener-Hollomon parameter as well as the W parameter of the Pivnik-Plata solid bonding criterion. The proposed model, developed for the AZ31 magnesium alloy using the commercial simulation package DEFORM, is 3D Lagrangian, thermo-mechanically coupled with visco-plastic material behavior. The model was first validated against experimental temperature measurements and then used to predi…
In-process control strategies for friction stir welding of AZ31 sheets with non-uniform thickness
2017
Two different in-process control strategies were developed and compared with the aim to produce AZ31 magnesium alloy joints by friction stir welding on sheet blanks with a non-uniform thickness. To this purpose, sheets with dip or hump zones were welded by either changing the rotational speed or the tool plunging in order to keep constant the value of the vertical force occurring during the welding stage of the process. The influence of the main process parameters on the tool force, the micro- and macromechanical properties, and the joints microstructures in the dip and hump zones were analyzed. The results showed that using the rotational speed change-based approach, the hump zones are sub…
Optimal control of discrete-time interval type-2 fuzzy-model-based systems with D-stability constraint and control saturation
2016
This paper investigates the optimal control problem for discrete-time interval type-2 (IT2) fuzzy systems with pole constraints. An IT2 fuzzy controller is characterized by two predefined functions, and the membership functions and the premise rules of the IT2 fuzzy controller can be chosen freely. The pole assignment is considered, which is constrained in a presented disk region. Based on Lyapunov stability theory, sufficient conditions of asymptotic stability with an H ∞ performance are obtained for the discrete-time IT2 fuzzy model based (FMB) system. Based on the criterion, the desired IT2 state-feedback controller is designed to guarantee that the closed-loop system is asymptotically s…