An Efficient Traffic Incident Detection and Classification Framework by Leveraging the Efficacy of Model Stacking

Table 3

Summary of reckless driving detection systems.

Ref #	ML technique	Independent variables	MFC
[37]	Supervised	Position, size, orientation, speed, reflect factor, and texture	✗
[38]	Supervised	Speed, position, acceleration, driver’s eye, and intoxication	✗
[39]	Supervised	Engine speed, battery voltage, clutch engagement, and TP	✗
[40]	Supervised	Image-based parameters, i.e., speed, acceleration, and angles	✗
[41]	Unsupervised	Position and velocity	✗
[42]	Supervised	Speed/acceleration	✗
[43]	Unsupervised	Speed and acceleration	✗
[44]	Supervised	Latitude, longitude, timestamp, and accelerations along the axis	✗