Virginia Tech Transportation Institute (VTTI)

Permanent URI for this community

https://hdl.handle.net/10919/5524

Browse

Browsing Virginia Tech Transportation Institute (VTTI) by Author "Abbott, A. Lynn"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Assessment of Psychophysiological Characteristics of Drivers Using Heart Rate from SHRP2 Face Video Data
    Sarkar, Abhijit; Doerzaph, Zachary R.; Abbott, A. Lynn (2014-08-25)
    The goal is to
    • Extract heart rate from face video
    • Understand the behavior of driver, e.g. cognitive load, panic attack, drowsiness, DUI
    • Develop automatic video reduction technique
    • Devise a tool for future
  • Thumbnail Image
    Non-contact Methods for Detecting Hot-mix Asphalt Nonuniformity
    de León Izeppi, Edgar (Virginia Tech, 2006-09-21)
    Segregation, or non-uniformity, in Hot Mix Asphalt (HMA) induces accelerated pavement distress(es) that can reduce a pavement's service life up to 50%. Quality Assurance procedures should detect and quantify the presence of this problem in newly constructed pavements. Current practices are usually based on visual inspections that identify non-uniform surface texture areas. An automatic process that reduces subjectivity would improve the quality-assurance procedures of HMA pavements. Virginia has undertaken a focused research effort to improve the uniformity of hot-mix asphalt (HMA) pavements. A method using a dynamic (laser-based) surface macrotexture instrument showed great promise, but it revealed that it may actually miss significant segregated areas because they only measure very thin longitudinal lines. The main objective of this research is to develop a non-contact system for the detection of segregated HMA areas and for the identification of the locations of these areas along a road for HMA quality assurance purposes. The developed system uses relatively low cost components and innovative image processing and analysis software. It computes the gray level co-occurrence matrix (GLCM) of images of newly constructed pavements to find various parameters that are commonly used in visual texture analysis. Using principal component analysis to integrate multivariable data into a single classifier, Hotelling's T2 statistic, the system then creates a list of the location of possible nonuniformities that require closer inspection. Field evaluations of the system at the Virginia Smart Road proved that it is capable of discriminating between different pavement surfaces. Verification of the system was conducted through a series of field tests to evaluate the uniformity of newly constructed pavements. A total of 18 continuous road segments of recently paved roads were tested and analyzed with the system. Tables and plots to be used by inspection personnel in the field were developed. The results of these field tests confirmed the capability of the system to detect potential nonuniformities of recently completed pavements. The system proved its potential as a useful tool in the final inspection process.
  • Thumbnail Image
    Video Magnification to Detect Heart Rate for Drivers
    Sarkar, Abhijit; Doerzaph, Zachary R.; Abbott, A. Lynn (National Surface Transportation Safety Center for Excellence, 2017-11-13)
    Heart rate is a strong indicator of a person’s psychophysiological state. For this reason, many applications would benefit from the ability to measure noncontact heart rate. The present work describes a new procedure for estimating blood volume pulse from video of a person’s face, with an emphasis on real-life scenarios like driving. The approach builds on the algorithm known as Eulerian VidMag, which has shown promise under laboratory conditions, but exhibits problems when applied in naturalistic situations. In particular, problems arise due to movement by the subject, changing illumination conditions, and low-frame-rate video. This work describes methods developed to address some of these problems, including working with video rates down to 10 frames per second. The methods were tested using videos of indoor subjects, as well as videos of drivers in naturalistic situations. We assessed the method through analysis of different stress levels using the extracted heart rate information for a driver on the road by comparing heart rate variability at different stress levels. Experiments showed that a systematic post-processing strategy can improve the accuracy of the VidMag algorithm’s raw output and achieve a good correlation in both instantaneous heart rate and average heart rate with the ground truth heart rate measurements. This, in particular, improves with higher quality video sources and more controlled experimental conditions. However, the robust broad application of the methods on existing lower-quality naturalistic video data remains a challenge.