Paper Title
Pothole Detection And Volume Estimation Using Stereoscopic Cameras

Pavement condition assessment is essential when developing road network maintenance programs. In practice, the data collection process is to a large extent automated. However, pavement distress detection (cracks, potholes, etc.) is mostly performed manually, which is labour-intensive and time-consuming. Existing methods either come along with high equipment and computation costs, or make use of acceleration data, which can only provide preliminary and rough condition surveys. In this paper, we present a method for automated pothole detection in asphalt pavement images. In the proposed method, an image is first resized and converted to grayscale. It is then histogram equalized. This is followed by thresholding and basic edge detection using the Sobel filter. Morphological operations such as dilation and eroding are performed and median filtering removes excess noise. A logical addition of the results obtained till then, results in a relative pothole shape. A structuring element allows for the pothole to be defined and opened and a skeleton of the pothole is formed. A bifurcation process is applied so as to remove any branches. Subsequently, the result of this operation is used for multiplying image matrices and the pothole is extracted. This methodology has been implemented in a MATLAB prototype and tested on 24 pavement images. The stereoscopic camera technique is used to improve accuracy and to determine the depth or volume of the pothole. The volume of the pothole may be used to estimate the amount of asphalt material required to fill it. The results indicate that this method can detect potholes in asphalt pavement images with reasonable accuracy. Keywords— Asphalt, Bifurcation, Grayscale, Median Filtering, Morphological Operations, Sobel