Bad conditions of road due to the potholes are one of the major cause of road damage and accidents to vehicles. Recently, with the increase in pollution and vehicular traffic, most of roads are being filled with many small and large potholes in most of places in the country. Detecting
potholes manually is a time-consuming task and labour-intensive task, automating this process which saves a lot of time and money. Hence, Many different methodologies have been implemented that is from reporting to authorities manually to the use of laser imaging. Though all of these techniques
have some disadvantages like risk while detection, high setup cost. By using the concept of Convolu-tional neural networks (CNN), a computer vision-based method we easily can identify the limitations, using the concept of Neural network by processing the image and detecting the potholes saves
a lot of time and money. Inputting the data to the model with the camera regularly to identify properly potholes, cracks. This is best options for automating bad-road identification problem.
Application of simplified convolutional neural networks for initial stator winding fault detection of the PMSM drive using different raw signal data