Visual Odometry Based on Omnidirectional Images
نویسندگان
چکیده
We propose a method for visual odometry using optical flow with a single omnidirectional (catadioptric) camera. We show how omnidirectional images can be used to perform optical flow, discussing the basis of optical flow and some restrictions needed to it and how unwarp these images. In special we describe how to unwarp omnidirectional images to Bird’s eye view, that correspond to scaled orthographic views of the ground plane. Catadioptric images facilitate landmark based odometry, since landmarks remain visible for longer time, as opposed to a small field-of-view standard camera. Also, providing adequate representations to support visual odometry with a fast processing time. We perform tests to measure robustness and performance of our approach with analysis of the data acquired. Keywords— Visual Odometry, Optical Flow, Catadioptric Images. Resumo— Neste trabalho propomos um método de odometria visual, empregando fluxo óptico, com um imagens omnidirecionais. Mostramos que imagens omnidirecionais podem ser usadas para realizar fluxo óptico, discutindo fluxo óptico e algumas restrições necessárias para realizar tal tarefa, em especial como retificar imagens catadiópticas para um tipo específico de vista denominado Bird’s eye view que corresponde a uma escala de visão ortográfica do terreno. Imagens omnidirecionais facilitam a odometria baseada em marcos, uma vez que estes marcos se mantém visiveis por mais tempo neste tipo de imagem, diferentemente de cameras padrões que apresentam um pequeno campo de visão. Além disso, este tipo de imagem fornece uma representação adequada para odometria visual com baixo custo computacional. Testes foram realizados para medir a robustez e desempenho da técnica desenvolvida com relação aos dados adquiridos. Keywords— Odometria Visual, Fluxo óptico, Imagens Catadiopticas
منابع مشابه
Correcting Vehicle Heading in Visual Odometry by Using Image Appearance
This paper describes a visual odometry algorithm for onroad vehicles under planar motion assumption. The algorithm uses as only input images taken by a single omnidirectional camera mounted on the roof of the vehicle. To achieve accurate motion estimation results, we decouple the rotation and translation estimation. In particular, we use a feature based ground plane tracker to estimate the tran...
متن کاملSpherical Image Processing for Accurate Visual Odometry with Omnidirectional Cameras
Due to their omnidirectional view, the use of catadioptric cameras is of great interest for robot localization and visual servoing. For simplicity, most vision-based algorithms use image processing tools (e.g. image smoothing) that were designed for perspective cameras. This can be a good approximation when the camera displacement is small with respect to the distance to the observed environmen...
متن کاملImproved Omnidirectional Odometry for a View-Based Mapping Approach
This work presents an improved visual odometry using omnidirectional images. The main purpose is to generate a reliable prior input which enhances the SLAM (Simultaneous Localization and Mapping) estimation tasks within the framework of navigation in mobile robotics, in detriment of the internal odometry data. Generally, standard SLAM approaches extensively use data such as the main prior input...
متن کاملMonocular Omnidirectional Visual Odometry for Outdoor Ground Vehicles
This paper describes an algorithm for visually computing the ego-motion of a vehicle relative to the road under the assumption of planar motion. The algorithm uses only images taken by a single omnidirectional camera mounted on the roof of the vehicle. The front ends of the system are two different trackers. The first one is a homography-based tracker that detects and matches robust scale invar...
متن کاملAn Application of Omnidirectional Vision to Grid-based SLAM in Indoor Environments
In this work we study the use of an omnidirectional camera for the estimation of a consistent metric representation of an indoor scenario. The proposed approach is based on Rao-Blackwellized Particle Filters and allows the robot to use a single vision sensor for estimating an occupancy grid map of the environment. The prediction phase of the filter is performed by means of an accurate visual od...
متن کامل