Camera Based Localization of a Smartphone in a Vehicle

Abstract

Open Access Research articleand Prevention (CDC), in year 2011 alone, 3,331 people were killed in accidents involving a distracted driver [18]. CDC also reports that around 31% of the drivers were reported to have used the text message or email features while driving. Combining these data, around 1,033 people were killed by texting and driving.Even without the statistical data, it is evident that cell phone usage while driving is a major distraction. The consequences of distracted driving are well recognized. In United States, as of March 2016, 14 states have laws that ban driving while talking on a handheld cell phone, and Forty-six States, ban text messaging for all drivers [2].Detecting vehicular movement is straightforward but ascertaining whether the user is indeed the driver is a complex problem. The fundamental problem of distinguishing between a driver and passenger using a mobile phone is the critical input to enable numerous safeties and interface enhancements. As in prior work, we seek to determine the in-vehicle location of the phone and use that as a heuristic to determine whether the phone is used by the driver or passenger. It uses a fundamentally different sensing approach to determine this location.Determining the movement of a vehicle is simple in that the speed can be calculated by using data from the GPS. Based on the speed determined, after a preselected speed, the cell phone can be disabled; or at least functions such as texting, games, can be disabled. However, determining if the user of the cell phone is indeed the driver is a challenging step. For an application that limits Smartphone functionality for a driver to execute successfully, the cell phone should be disabled only when it is used by a person driving a vehicle.Determining position of a Smartphone in a vehicle can enable and provide context for numerous interesting in-vehicle applications, including driver distraction prevention, driving behaviour estimation, in-vehicle services customization, etc. However, most of the existing work on in-vehicle positioning relies on special infrastructures, such as the stereo, cigarette lighter adapter or OBD (on-board diagnostic) adapter. In this paper, we propose an infrastructure-free, in-vehicle positioning system via smart phones. Our methodology uses only embedded camera in smart phones to determine the phones' seat-level locations in a vehicle. AbstractDistracted driving is any activity that could divert a person's attention away from the primary task of driving. All distractions endanger driver, passenger, and bystander safety. Cell phone usage while driving is a major distraction. Multiple solutions have been proposed on limiting smart phone functionality when being used by a driver. However, it remains challenge to detect if it is indeed the driver using the Smartphone.The objective of this paper is to address the fundamental problem of distinguishing the driver from passengers, non-intrusively, i.e., enabling driver phone detection without relying on any particular situations, events...