An effective framework to detect the lane border using convolutional neural network over K-means clustering.

A CNN can automatically detect lane borders using k-means clustering. This 10-node By combining a convolutional neural network with the k-means clustering approach, an automated system was developed for lane line identification. The following G-power parameters were used to split the training dataset in half and the testing dataset in thirds: α=0.05 and power=0.85. After correcting for any confounding variables, the 84% accuracy of k-means clustering and the 94% accuracy of the convolutional neural network (CNN) (p>0.05, t=0.430). When compared to clustering methods, a Convolutional Neural Network was more effective in classifying Lane Line. [ABSTRACT FROM AUTHOR]