Search Results

The driver's face detection and tracking method important for Advanced Driver Assistance Systems (ADAS) and autonomous driving in various situations. The deep SORT algorithm has integrated appearance information, the motion model and the intersection-over-union (IOU) distance methods, and has been applied to face tracking, but it depends on detection information in every frame. Once the detection information lacks, the deep SORT algorithm will wait until the target detects bounding boxes appear again, even if the target didn’t disappear or shield. Hence, we propose to use a new tracker that not completely depend on the detection algorithm to cascade with the deep SORT algorithm to realize stable driver's face tracking. At first, the driver's face detection and tracking will be accomplished by the MTCNN-deep-SORT algorithm.

The research question investigated in this study is what are the key attributes of foot and ankle injury in the between-rail frontal crash? For the foot and ankle, what was the type of interior surface contacted and the type of resulting trauma? The method was to study with in-depth case reviews of NASS-CDS cases where a driver suffered an AIS=2 foot or ankle injury in between-rail crashes. Cases were limited to belted occupants in vehicles equipped with air bags. The reviews concentrated on coded and non-coded data, identifying especially those factors contributing to the injuries of the driver's foot/ankle. This study examines real-world crash data between the years 1997-2009 with a focus on frontal crashes involving 1997 and later model year vehicles. The raw data count for between-rail crashes was 732, corresponding to 227,305 weighted, tow-away crashes.

Previously, a 10-year-old (YO) pediatric thorax finite element model (FEM) was developed and verified against child chest stiffness data measured from clinical cardiopulmonary resuscitation (CPR). However, the CPR experiments were performed at relatively low speeds, with a maximum loading rate of 250 mm/s. Studies showed that the biomechanical responses of human thorax exhibited rate sensitive characteristics. As such, the studies of dynamic responses of the pediatric thorax FEM are needed. Experimental pediatric cadaver data in frontal pendulum impacts and diagonal belt dynamic loading tests were used for dynamic validation. Thoracic force-deflection curves between test and simulation were compared. Strains predicted by the FEM and the injuries observed in the cadaver tests were also compared for injury assessment and analysis. This study helped to further improve the 10 YO pediatric thorax FEM.