Overview of Performance and Accuracy of Smartphone Sensors in Augmented Reality Applications
Author(s):
Abstract:
Since incorrect excavations have resulted in extensive and irreparable financial and physical losses, therefore different drillings require having accurate information about the status of the infrastructures. Ubiquitous Geospatial Information System (UBGIS) as a new generation of Geospatial Information System (GIS) can be a good solution to avoid such problems. Augmented Reality (AR) is the next generation of real-world 3D visualization that can be used to visualize this infrastructure on smart phones. These days, due to recent advances in the mobile phone hardware industry, such as increased processing power and the availability of various sensors such as GPS, accelerometer, gyroscope, compass, camera and mobile phone display, AR is easily available to developers and users. The exterior orientation parameters of camera are required for AR and tracking. These parameters can be achieved by sensors embedded in smart phones. So overview of type, performance and accuracy of sensors is critical. This paper aims to examine the performance and accuracy of the types of sensors in the smartphone in order to visualize the underground infrastructure using augmented reality technology. For this purpose, two smartphones Samsung Galaxy S4 and iPhone4 have been reviewed. The Samsung S4 also has less radial distortion than the iPhone 4 (about half) and it is possible to correct it linearly. The best accuracy of GPS is about 10-20 m. The average of compass error is about 10 degrees in flat regions and about 30 degrees in regions surrounding with high buildings. Gyroscope values drift about 3-4 degrees per second and increase gradually. Also heading values in regions surrounding with high buildings have less accuracy. Generally choosing the best sensor and smartphone depend on usage. Also Samsung Galaxy S9 and iPhone 8 as the new generation of these two operating systems are compared.
Keywords:
Language:
Persian
Published:
Geospatial Engineering Journal, Volume:10 Issue: 1, 2019
Pages:
49 to 62
https://magiran.com/p1954790