Table 2.

A comparative evaluation of the smart sensors used in RCM.

Key Variables	Camera	Laser	GPR	Thermal	Vibration
Technology	2D imaging	3D construction of image using reflection	Radio waves to explore underground surface; creates 3D image of sub-surface	Based on the change in temperature of surrounding objects using infrared waves	Accelerometers, gyroscope, and GPS readings
Processing	Complex image-processing algorithms	Collection of 3D point cloud	Collection of depth images and simulation data required	Collection of heat variation of surface	Readings are directly used
Real-Time Application	Processor dependent	Yes	Yes	Yes	Cannot be used in real-time detection
Sensing Time	While approaching distress	While approaching distress	While approaching distress	While approaching distress	Only after experiencing distress
Characterization of Distress	Based on shape and size	Based on 3D image	Based on 3D image	Based on heat maps	Detection only along wheel path as 1D parameters
Light Sensitivity	Sensitive to illuminance levels, light source position	Not sensitive to light effect	Not sensitive to light effect	Not sensitive to light effect, but surface temperatures	None
Accuracy	Algorithm dependent	High	High	High	Highly susceptible to errors
Resolution	Varying low to high	High-resolution images	Depends on frequency	Needs improvement	-
Processing Time	Data collection and analysis is fast; response time is processor dependent	Data collection is fast and can be collected at speeds as high as 100 km/h	Delayed due to large data processing; however, data collection is automated	Data collection and analysis is fast	Poor as data processing is required
Cost	Economical	High	Highly expensive	Very expensive	Low
Data Type	2D, 3D	3D	3D	2D, 3D	1D