Vision-based Heading Misalignment Estimation for Portable Devices Using Head/Shoulder Detection

Abdelrahman Ali, Amr Al-Hamad, Jacques Georgy, and Chris Goodall

Abstract:	Recently, most of smart devices such as smartphone are released with front and rare cameras which are commonly used in daily life activities and involved in many applications. During the time that users read, text, or even take notes on their devices; the camera can be used independently to provide information about the direction of the device with respect to the user direction. Therefore, exploiting the extracted information from the captured images can improve the pedestrian navigation by determining and compensating for the heading misalignment angle between the user and the portable navigation device. The proposed algorithm used the captured image/sequence of images from the smart device camera that faces the user’s body to estimate the heading misalignment angle. The captured images that contain user’s face and shoulders regions are picked. Then, an edge detection technique is used to provide the outer borders of the person’s face and shoulders. The relation between the captured image and the device orientation is used to estimate the required angle. The technique covers a wide range of different portable navigation device orientations. This is done by identifying the user face, shoulders, or a combination to determine the heading misalignment angle. This work is patent pending. The information from these images can be integrated with misalignment estimation techniques based on the smart device’s on-board sensors. Adding another source of heading misalignment angle estimation by using the image information enables a better estimate of the misalignment angle, which in turn, improves the navigation solution. Different test scenarios were conducted to assess the performance of the proposed technique. The results show that the proposed technique is able to estimate the correct heading misalignment angle.
Published in:	Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015) September 14 - 18, 2015 Tampa Convention Center Tampa, Florida
Pages:	67 - 73
Cite this article:	Ali, Abdelrahman, Al-Hamad, Amr, Georgy, Jacques, Goodall, Chris, "Vision-based Heading Misalignment Estimation for Portable Devices Using Head/Shoulder Detection," Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015), Tampa, Florida, September 2015, pp. 67-73.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In