In the modern era, high-speed vehicles are emerging out and driver want to run them at the full designed speed. But sometimes, the driver is not able to do so because of potholes, cracks, and other bad road conditions. Further, these factors are also responsible for wear and tear of the automobile, inconvenience of passengers, more fuel consumption of fuel, and as well as loss of human life in road accidents. So, detection of bad road conditions, mainly potholes, is very important to improve and maintain the road. Several image processing approaches were implemented to automatic monitoring the pavement surface to detect the potholes. But various road conditions and different scale, shape, size, and illumination effect of potholes led to unacceptable stability of approaches. Therefore, in this paper, convolution neural network is used to automatically detect and analyses the road conditions using digital images. It determines the very precise and accurate depth, area, and shape from digital images.
A Vision-Based Pothole Detection Using CNN Model
SN COMPUT. SCI.
SN Computer Science ; 4 , 6
2023-09-23
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Pothole , Feature extraction , Convolution neural network (CNN) , Residual networks Computer Science , Computer Science, general , Computer Systems Organization and Communication Networks , Software Engineering/Programming and Operating Systems , Data Structures and Information Theory , Information Systems and Communication Service , Computer Imaging, Vision, Pattern Recognition and Graphics
Pothole Detection Using Computer Vision and Learning
IEEE | 2020
|Road Pothole Detection System Based on Stereo Vision
IEEE | 2018
|