Abstract
Potholes are the most common form of distress on cement concrete pavements, which can compromise pavement safety and ridability. Thus, timely and accurate pothole detection is an important task in developing proper maintenance strategies and ensuring driving safety. This paper proposes a method of integrating the processing of grayscale and texture features. This method mainly combines industrial camera to realize rapid and accurate detection of pothole. Image processing techniques including texture filters, image grayscale, morphology, and extraction of the maximum connected domain are used synergistically to extract useful features from digital images. A machine learning model based on the library for support vector machine (LIBSVM) is constructed to distinguish potholes from longitudinal cracks, transverse cracks, and complex cracks. The method is validated using data collected from agricultural and pastoral areas of Inner Mongolia, China. The comprehensive experiments for recognition of potholes show that the recall, precision, and F1-Score achieved are 100%, 97.4%, and 98.7%, respectively. In addition, the overlap rate between the extracted pothole region and the original image is estimated. Images with an overlap rate greater than 90% accounted for 76.8% of the total image, and images with an overlap rate greater than 80% accounted for 94% of the total image. A comparison discloses that the proposed approach is superior to the existing method not only from the perspective of the accuracy of pothole detection but also from the perspective of the segmentation effect and processing efficiency.
Highlights
Cement concrete pavement, as one of the important parts of a road network, has higher strength, better stability, longer service life, and lower maintenance cost
According to the collected 610 defect images, the sample types were labeled. e damage of the pavement aggregate exposed in the image is determined as a pothole, and there were 250 potholes examined . e crack that is nearly perpendicular to the driving direction is determined as a transverse crack, and there were 150 images examined
E sample label is defined by four types: class one is pothole, class two is transverse crack, class three is longitudinal crack, and class four is complex crack, respectively. e four types of samples have eight attributes, namely, the number of connected domains, projection ratio, fractal dimension, area, perimeter, circularity, rectangle degree, and aspect ratio
Summary
As one of the important parts of a road network, has higher strength, better stability, longer service life, and lower maintenance cost. E main aim of this study is to focus on cement concrete pavement, using image grayscale and texture features, combining with the LIBSVM classifier, and relying on simple acquisition equipment to realize fast, accurate, and low-cost detection and segmentation of potholes, providing a means and method for rapid road detection. Different features are extracted from each image to distinguish potholes from other distress types To this end, a machine learning algorithm is selected, and the extracted features are used as training data to construct a detection model. To ensure recognition accuracy when detecting pavement potholes, it is often necessary to distinguish other types of distress, so this paper faces the problem of multiclassification To meet such classification requirements, LIBSVM [56] is chosen as the SVM tool. Evaluation of the prediction effect of LIBSVM can be calculated by the following measures [56]: accuracy correctly predicted data × 100%. total testing data (18)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
