Validity and reliability of naturalistic driving scene categorization judgments from crowdsourcing

Christopher D. Cabrall
Zhenji Lu
Miltos Kyriakidis
Laura Manca
Chris Dijksterhuis
Riender Happee
Joost de Winter

A common challenge with processing naturalistic driving data for many different possible driving research interests or applications is that humans may need to categorize great volumes of recorded visual information until automated algorithms might be trained to do so alone.

This study, by means of the online platform CrowdFlower, investigated the potential of crowdsourcing to provide content identification categorizations of driving scene features (e.g., presence of another vehicle, straight road segments, etc.) at greater scale than a single person or a small team of researchers would be capable of. The validity and reliability of CrowdFlower results were examined, both with and without employing a set of randomly embedded controlled questions (Gold Test Questions) intermixed with experimental questions (Work Mode). In total, 200 workers from 46 countries participated in this study, and the collection of data lasted one and a half days.

By employing Gold Test Questions, we found significantly more accurate and consistent responses from external workers at both a smaller and larger scale of video segment categorizations for the identification of common driving scene elements (e.g., position and behavior of other vehicles, road and signage characteristics, etc.). In terms of validity and at the small scale, an average accuracy of 91% on paired items was found with the controlled questions compared to 78% without. A difference in bias was found where without Gold Test Questions external workers returned more false positives than false negatives whereas the opposite was found true of the condition with Gold Test Questions. At the large scale (making use of the controlled questions), a random subset of categorizations returned similar levels of accuracy (95%) and a similar pattern of error bias. In terms of reliability and at the small scale, where segments were rated in triplicate redundancy, the percentage of unanimous agreement was found significantly higher when using controlled questions (90%) than without them (65%). Across the small scale of internally validated answers, more than two-thirds of any correct categorization were unanimously returned and 86% or more of any correct categorization was returned by a majority vote. Where it would be infeasible to validate every response for accuracy, similar voting reliability results were found to exist across the responses of the large scale.

Overall results support compelling evidence for CrowdFlower as being able to yield valid and reliable crowdsourced categorizations of naturalistic driving scene contents in a short period of time and thus a potentially powerful and as-of-yet under-utilized resource in the toolbox of driving research and driving automation development.

MEET US


7
Dec

Shipping and the environment – research meets reality

Centre for Transport Studies (CTS) in co-operation with Ports of Stockholm invite you to the seminar Shipping and the environment – research meets reality.

LATEST NEWS


2017-10-16

ERPUG Forum

The five-year anniversary of European Road Profile Users' Group (ERPUG) Forum will take place at Ramboll head quarter, Copenhagen, Denmark October 19-20, 2017.


2017-09-29

Self-driving buses in Sweden next year?

A self-driving, fossil-free bus. This idea might become reality through a forthcoming collaborative project involving the Swedish National Road and Transport Research Institute (VTI), Linköping University and several other participants. The project group aim...


2017-09-29

New climate-proof solutions for hard surfaces in cities

High-density road infrastructure that emphasise maximum durability and minimum maintenance create inflexible systems, which put increased stress on urban trees and lead to increased risk for flooding. Over the past five years, the ‘Climate-proof solutions for...


2017-09-29

VTI is preparing for automated vehicles

Automation of traffic systems will lead to major changes. The European Union’s (EU) CoEXist research project began in June 2017 with the aim of preparing cities and road operators for the introduction of self-driving vehicles. The Swedish National Road and...


2017-07-05

Vehicle Driver Monitoring: sleepiness and cognitive load

To prevent road crashes it is important to understand driver related contributing factors, which have been suggested to be the critical reason in 94 per cent of crashes. The overall aim of the project Vehicle Driver Monitoring has been to advance the...


2017-06-29

Tomas Svensson new director-general

Tomas Svensson was today appointed Director-General of VTI. Tomas has been acting Director-General since January 2017.