The accuracy dependency investigation of simultaneous localization and mapping on the errors from mobile device sensors

Abstract

Monocular Simultaneous Localization and Mapping (SLAM) is one of the most complex and well-known problems, affecting several scientific fields: robotics, computer vision, virtual reality. This paper aims to study the SLAM problem for the mobile device with a monocular camera and sensors: accelerometer, gyroscope and digital compass. The latter allow to obtain an additional estimation of a mobile device position and orientation. The aim is to assess the potential suitability and efficiency of using extra information from inertial sensors to improve the solution quality and to reduce the time to obtain the solution. The experimental part of the study, including both model and field experiments, allowed to determine the requirements for permissible errors introduced by the sensors of the mobile device. For a specific model of a mobile device, it is shown that the electronic compass meets these requirements, while the errors of the inertial sensors used to determine the movements are unacceptably large.

Weighted combination of per-frame recognition results for text recognition in a video stream

“…Also, for a better reconstruction of a recognized document image, it is possible to use the data obtained from various sensors of the recording device, such as, for example, an accelerometer or a gyroscope. However, for modern mobile devices, the error in their measurements can be quite significant and prevent using this data for image reconstruction [43]. The second group of methods involves combining the results of individual image recognition.…”

The Influence of Camera and Optical System Parameters on the Uncertainty of Object Location Measurement in Vision Systems

“…There are many works on vision systems or camera systems that assess the measurement uncertainty: mechanics, machine-vision [ 30 , 31 , 32 ], flow measurements [ 33 , 34 , 35 ], geoinformation sciences [ 36 , 37 , 38 ], or medicine [ 39 ]. Naturally, in each of the presented examples, the authors performed estimates of quality parameters, most often as measurement errors or standard deviations.…”